Publications

@inproceedings{ERK_Marija_2021,

title = {A Comparative Study on Discriminative and One--Class Learning Models for Deepfake Detection},

author = {Marija Ivanovska and Vitomir Štruc},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2021/10/ERK_2021__A_Comparative_Study_of_Discriminative_and_One__Class_Learning_Models_for_Deepfake_Detection.pdf},

year  = {2021},

date = {2021-09-20},

booktitle = {Proceedings of ERK 2021},

pages = {1--4},

abstract = {Deepfakes or manipulated face images, where a donor's face is swapped with the face of a target person, have gained enormous popularity among the general public recently. With the advancements in  artificial intelligence and generative modeling  

such images can nowadays be easily generated and used to spread misinformation and harm individuals, businesses or society. As the tools for generating deepfakes are rapidly improving, it is critical for deepfake detection models to be able to recognize advanced, sophisticated data manipulations, including those that have not been seen during  training. In this paper, we explore the use of one--class learning models as an alternative to discriminative methods for the detection of deepfakes. We conduct a  comparative study with three popular deepfake datasets and investigate the performance of selected (discriminative and one-class) detection models in matched- and cross-dataset experiments. Our results show that disciminative models significantly outperform one-class models when training and testing data come from the same dataset, but degrade considerably when the characteristics of the testing data deviate from the training setting. In such cases, one-class models tend to generalize much better.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Grm-SuperResolution_ERK2021,

title = {Frequency Band Encoding for Face Super-Resolution},

author = {Klemen Grm and Štruc Vitomir},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2021/10/SRAE_ERK21.pdf},

year  = {2021},

date = {2021-09-10},

booktitle = {Proceedings of ERK 2021},

pages = {1-4},

abstract = {In this paper, we present a novel method for face super-resolution based on an encoder-decoder architecture. Unlike previous approaches, which focused primarily on directly reconstructing the high-resolution face appearance from low-resolution images, our method relies on a multi-stage approach where we learn a face representation in different frequency bands, followed by decoding the representation into a high-resolution image. Using quantitative experiments, we are able to demonstrate that this approach results in better face image reconstruction, as well as aiding in downstream semantic tasks such as face recognition and face verification.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{NIR_IJCB2021,

title = {NIR Iris Challenge Evaluation in Non-cooperative Environments: Segmentation and Localization},

author = {Caiyong Wang and Yunlong Wang and Kunbo Zhang and Jawad Muhammad and Tianhao Lu and Qi Zhang and Qichuan Tian and Zhaofeng He and Zhenan Sun and Yiwen Zhang and Tianbao Liu and Wei Yang and Dongliang Wu and Yingfeng Liu and Ruiye Zhou and Huihai Wu and Hao Zhang and Junbao Wang and Jiayi Wang and Wantong Xiong and Xueyu Shi and Shao Zeng and Peihua Li and Haodong Sun and Jing Wang and Jiale Zhang and Qi Wang and Huijie Wu and Xinhui Zhang and Haiqing Li and Yu Chen and Liang Chen and Menghan Zhang and Ye Sun and Zhiyong Zhou and Fadi Boutros and Naser Damer and Arjan Kuijper and Juan Tapia and Andres Valenzuela and Christoph Busch and Gourav Gupta and Kiran Raja and Xi Wu and Xiaojie Li and Jingfu Yang and Hongyan Jing and Xin Wang and Bin Kong and Youbing Yin and Qi Song and Siwei Lyu and Shu Hu and Leon Premk and Matej Vitek and Vitomir Štruc and Peter Peer and Jalil Nourmohammadi Khiarak and Farhang Jaryani and Samaneh Salehi Nasab and Seyed Naeim Moafinejad and Yasin Amini and Morteza Noshad},

url = {https://ieeexplore.ieee.org/iel7/9484326/9484328/09484336.pdf?casa_token=FOKx4ltO-hYAAAAA:dCkNHfumDzPGkAipRdbppNWpzAiUYUrJL6OrAjNmimTxUA0Vmx311-3-J3ej7YQc_zONxEO-XKo},

doi = {10.1109/IJCB52358.2021.9484336},

year  = {2021},

date = {2021-08-01},

booktitle = {Proceedings of the IEEE International Joint Conference on Biometrics (IJCB 2021)},

abstract = {For iris recognition in non-cooperative environments, iris segmentation has been regarded as the first most important challenge still open to the biometric community, affecting all downstream tasks from normalization to recognition. In recent years, deep learning technologies have gained significant popularity among various computer vision tasks and also been introduced in iris biometrics, especially iris segmentation. To investigate recent developments and attract more interest of researchers in the iris segmentation method, we organized the 2021 NIR Iris Challenge Evaluation in Non-cooperative Environments: Segmentation and Localization (NIR-ISL 2021) at the 2021 International Joint Conference on Biometrics (IJCB 2021). The challenge was used as a public platform to assess the performance of iris segmentation and localization methods on Asian and African NIR iris images captured in non-cooperative environments. The three best-performing entries achieved solid and satisfactory iris segmentation and localization results in most cases, and their code and models have been made publicly available for reproducibility research.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{MFR_IJCB2021,

title = {MFR 2021: Masked Face Recognition Competition},

author = {Fadi Boutros and Naser Damer and Jan Niklas Kolf and Kiran Raja and Florian Kirchbuchner and Raghavendra Ramachandra and Arjan Kuijper and Pengcheng Fang and Chao Zhang and Fei Wang and David Montero and Naiara Aginako and Basilio Sierra and Marcos Nieto and Mustafa Ekrem Erakin and Ugur Demir and Hazım Kemal Ekenel and Asaki Kataoka and Kohei Ichikawa and Shizuma Kubo and Jie Zhang and Mingjie He and Dan Han and Shiguang Shan and Klemen Grm and Vitomir Štruc and Sachith Seneviratne and Nuran Kasthuriarachchi and Sanka Rasnayaka and Pedro C. Neto and Ana F. Sequeira and Joao Ribeiro Pinto and Mohsen Saffari and Jaime S. Cardoso},

url = {https://ieeexplore.ieee.org/iel7/9484326/9484328/09484337.pdf?casa_token=OOL4s274P0YAAAAA:XE7ga2rP_wNom2Zeva75ZwNwN-HKz6kF1HZtkpzrdTdz36eaGcLffWkzOgIe3xU2PqaU30qTLws},

doi = {10.1109/IJCB52358.2021.9484337},

year  = {2021},

date = {2021-08-01},

booktitle = {Proceedings of the IEEE International Joint Conference on Biometrics (IJCB 2021)},

abstract = {This paper presents a summary of the Masked Face Recognition Competitions (MFR) held within the 2021 International Joint Conference on Biometrics (IJCB 2021). The competition attracted a total of 10 participating teams with valid submissions. The affiliations of these teams are diverse and associated with academia and industry in nine different countries. These teams successfully submitted 18 valid solutions. The competition is designed to motivate solutions aiming at enhancing the face recognition accuracy of masked faces. Moreover, the competition considered the deployability of the proposed solutions by taking the compactness of the face recognition models into account. A private dataset representing a collaborative, multisession, real masked, capture scenario is used to evaluate the submitted solutions. In comparison to one of the topperforming academic face recognition solutions, 10 out of the 18 submitted solutions did score higher masked face verification accuracy.

},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{BortolatoFG2020,

title = {Learning privacy-enhancing face representations through feature disentanglement},

author = {Blaž Bortolato and Marija Ivanovska and Peter Rot and Janez Križaj and Philipp Terhorst and Naser Damer and Peter Peer and Vitomir Štruc},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2020/07/FG2020___Learning_privacy_enhancing_face_representations_through_feature_disentanglement-1.pdf

},

year  = {2020},

date = {2020-11-04},

booktitle = {Proceedings of FG 2020},

publisher = {IEEE},

abstract = {Convolutional Neural Networks (CNNs) are today the de-facto standard for extracting compact and discriminative face representations (templates) from images in automatic face recognition systems. Due to the characteristics of CNN models, the generated representations typically encode a multitude of information ranging from identity to soft-biometric attributes, such as age, gender or ethnicity. However, since these representations were computed for the purpose of identity recognition only, the soft-biometric information contained in the templates represents a serious privacy risk. To mitigate this problem, we present in this paper a privacy-enhancing approach capable of suppressing potentially sensitive soft-biometric information in face representations without significantly compromising identity information. Specifically, we introduce a Privacy-Enhancing Face-Representation learning Network (PFRNet) that disentangles identity from attribute information in face representations and consequently allows to efficiently suppress soft-biometrics in face templates. We demonstrate the feasibility of PFRNet on the problem of gender suppression and show through rigorous experiments on the CelebA, Labeled Faces in the Wild (LFW) and Adience datasets that the proposed disentanglement-based approach is highly effective and improves significantly on the existing state-of-the-art.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{SSBC2020,

title = {SSBC 2020: Sclera Segmentation Benchmarking Competition in the Mobile Environment},

author = {M. Vitek and A. Das and Y. Pourcenoux and A. Missler and C. Paumier and S. Das and I. De Ghosh and D. R. Lucio and L. A. Zanlorensi Jr. and D. Menotti and F. Boutros and N. Damer and J. H. Grebe and A. Kuijper and J. Hu and Y. He and C. Wang and H. Liu and Y. Wang and Z. Sun and D. Osorio-Roig and C. Rathgeb and C. Busch and J. Tapia and A.~Valenzuela and G. Zampoukis and L. Tsochatzidis and I. Pratikakis and S. Nathan and R. Suganya and V. Mehta and A. Dhall and K. Raja and G. Gupta and J. N. Khiarak and M. Akbari-Shahper and F. Jaryani and M. Asgari-Chenaghlu and R. Vyas and S. Dakshit and S. Dakshit and P. Peer and U. Pal and V. Štruc},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2020/11/IJCB_SSBC_2020.pdf},

year  = {2020},

date = {2020-09-28},

booktitle = {International Joint Conference on Biometrics (IJCB 2020)},

pages = {1--10},

abstract = {The paper presents a summary of the 2020 Sclera Segmentation Benchmarking Competition (SSBC), the 7th in the series of group benchmarking efforts centred around the problem of sclera segmentation. Different from previous editions, the goal of SSBC 2020 was to evaluate the performance of sclera-segmentation models on images captured with mobile devices. The competition was used as a platform to assess the sensitivity of existing models to i) differences in mobile devices used for image capture and ii) changes in the ambient acquisition conditions. 26 research groups registered for SSBC 2020, out of which 13 took part in the final round and submitted a total of 16 segmentation models for scoring. These included a wide variety of deep-learning solutions as well as one approach based on standard image processing techniques. Experiments were conducted with three recent datasets. Most of the segmentation models achieved relatively consistent  performance across images captured with different mobile devices (with slight differences across devices), but struggled most with  low-quality images captured in challenging ambient conditions, i.e., in an indoor environment and with poor lighting. },

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Biosig_naser_2020,

title = {Privacy Evaluation Protocols for the Evaluation of Soft-Biometric Privacy-Enhancing Technologies},

author = {Philipp Terhörst, Marco Huber, Naser Damer, Peter Rot, Florian Kirchbuchner, Vitomir Struc, Arjan Kuijper},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2020/11/Biosig_privacy.pdf},

issn = {1617-5468},

year  = {2020},

date = {2020-09-16},

booktitle = {Proceedings of the International Conference of the Biometrics Special Interest Group (BIOSIG) 2020},

pages = {1-5},

publisher = {IEEE},

abstract = {Biometric data includes privacy-sensitive information, such as soft-biometrics. Soft-biometric privacy enhancing technologies aim at limiting the possibility of deducing such information. Previous works proposed several solutions to this problem using several different evaluation processes, metrics, and attack scenarios. The absence of a standardized evaluation protocol makes a meaningful comparison of these solutions difficult. In this work, we propose privacy evaluation protocols (PEPs) for privacy-enhancing technologies (PETs) dealing with soft-biometric privacy. Our framework evaluates PETs in the most critical scenario of an attacker that knows and adapts to the systems privacy-mechanism. Moreover, our PEPs differentiate between PET of learning-based or training-free nature. To ensure that our protocol meets the highest standards in both cases, it is based on Kerckhoffs‘s principle of cryptography.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{GrmEUSIPCO2020,

title = {Analysis of Race and Gender Bias in Deep Age Estimation Model},

author = {Andraž Puc and Vitomir Štruc and Klemen Grm},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2020/07/race_and_gender_bias_eusipco-2.pdf},

year  = {2020},

date = {2020-09-01},

booktitle = {Proceedings of EUSIPCO 2020},

abstract = {Due to advances in deep learning and convolutional neural networks (CNNs) there has been significant progress in the field of visual age estimation from face images over recent years. While today's models are able to achieve considerable age estimation accuracy, their behaviour, especially with respect to specific demographic groups is still not well understood. In this paper, we take a deeper look at CNN-based age estimation models and analyze their performance across different race and gender groups. We use two publicly available off-the-shelf age estimation models, i.e., FaceNet and WideResNet, for our study and analyze their performance on the UTKFace and APPA-REAL datasets. We partition face images into sub-groups based on race, gender and combinations of race and gender. We then compare age estimation results and find that there are noticeable differences in performance across demographics. Specifically, our results show that age estimation accuracy is consistently higher for men than for women, while race does not appear to have consistent effects on the tested models across different test datasets.

},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{sircelj2020sqcnn,

title = {Segmentation and Recovery of Superquadric Models using Convolutional Neural Networks},

author = {Jaka Šircelj and Tim Oblak and Klemen Grm and Uroš Petković and Aleš Jaklič and Peter Peer and Vitomir Štruc and Franc Solina},

url = {https://lmi.fe.uni-lj.si/en/sircelj2020cvww/

https://arxiv.org/abs/2001.10504},

year  = {2020},

date = {2020-02-03},

booktitle = {25th Computer Vision Winter Workshop (CVWW 2020)},

abstract = {In this paper we address the problem of representing 3D visual data with parameterized volumetric shape primitives. Specifically, we present a (two-stage) approach built around convolutional neural networks (CNNs) capable of segmenting complex depth scenes into the simpler geometric structures that can be represented with superquadric models. In the first stage, our approach uses a Mask RCNN model to identify superquadric-like structures in depth scenes and then fits superquadric models to the segmented structures using a specially designed CNN regressor. Using our approach we are able to describe complex structures with a small number of interpretable parameters. We evaluated the proposed approach on synthetic as well as real-world depth data and show that our solution does not only result in competitive performance in comparison to the state-of-the-art, but is able to decompose scenes into a number of superquadric models at a fraction of the time required by competing approaches. We make all data and models used in the paper available from https://lmi.fe.uni-lj.si/en/research/resources/sq-seg.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{oblak2019recovery,

title = {Recovery of Superquadrics from Range Images using Deep Learning: A Preliminary Study},

author = {Tim Oblak and Klemen Grm and Aleš Jaklič and Peter Peer and Vitomir Štruc and Franc Solina},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/Superkvadriki_draft.pdf},

year  = {2019},

date = {2019-06-01},

booktitle = {2019 IEEE International Work Conference on Bioinspired Intelligence (IWOBI)},

journal = {arXiv preprint arXiv:1904.06585},

pages = {45-52},

publisher = {IEEE},

abstract = {It has been a longstanding goal in computer vision to describe the 3D physical space in terms of parameterized volumetric models that would allow autonomous machines to understand and interact with their surroundings. Such models are typically motivated by human visual perception and aim to represents all elements of the physical word ranging from individual objects to complex scenes using a small set of parameters. One of the de facto standards to approach this problem are superquadrics - volumetric models that define various 3D shape primitives and can be fitted to actual 3D data (either in the form of point clouds or range images). However, existing solutions to superquadric recovery involve costly iterative fitting procedures, which limit the applicability of such techniques in practice. To alleviate this problem, we explore in this paper the possibility to recover superquadrics from range images without time consuming iterative parameter estimation techniques by using contemporary deep-learning models, more specifically, convolutional neural networks (CNNs). We pose the superquadric recovery problem as a regression task and develop a CNN regressor that is able to estimate the parameters of a superquadric model from a given range image. We train the regressor on a large set of synthetic range images, each containing a single (unrotated) superquadric shape and evaluate the learned model in comparative experiments with the current state-of-the-art. Additionally, we also present a qualitative analysis involving a dataset of real-world objects. The results of our experiments show that the proposed regressor not only outperforms the existing state-of-the-art, but also ensures a 270x faster  execution time.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{emervsivc2019unconstrained,

title = {The Unconstrained Ear Recognition Challenge 2019},

author = {Žiga Emeršič and A. Kumar S. V. and B. S. Harish and W. Gutfeter and J. N. Khiarak and A. Pacut and E. Hansley and M. Pamplona Segundo and S. Sarkar and H. Park and G. Pyo Nam and I. J. Kim and S.G. Sangodkar and U. Kacar and M. Kirci and L. Yuan and J. Yuan and H. Zhao and F. Lu and J. Mao and X. Zhang and D. Yaman and F. I. Eyiokur and K. B. Ozler and H. K. Ekenel and D. Paul Chowdhury and S. Bakshi and P. K. Sa and B. Majhni and P. Peer and V. Štruc},

url = {https://arxiv.org/pdf/1903.04143.pdf},

year  = {2019},

date = {2019-06-01},

booktitle = {International Conference on Biometrics (ICB 2019)},

journal = {arXiv preprint arXiv:1903.04143},

abstract = {This paper presents a summary of the 2019 Unconstrained Ear Recognition Challenge (UERC), the second in a series of group benchmarking efforts centered around the problem of person recognition from ear images captured in uncontrolled settings. The goal of the challenge is to assess the performance of existing ear recognition techniques on a challenging large-scale ear dataset and to analyze performance of the technology from various viewpoints, such as generalization abilities to unseen data characteristics, sensitivity to rotations, occlusions and image resolution and performance bias on sub-groups of subjects, selected based on demographic criteria, i.e. gender and ethnicity. Research groups from 12 institutions entered the competition and submitted a total of 13 recognition approaches ranging from descriptor-based methods to deep-learning models. The majority of submissions focused on ensemble based methods combining either representations from multiple deep models or hand-crafted with learned image descriptors. Our analysis shows that methods incorporating deep learning models clearly outperform techniques relying solely on hand-crafted descriptors, even though both groups of techniques exhibit similar behaviour when it comes to robustness to various covariates, such presence of occlusions, changes in (head) pose, or variability in image resolution. The results of the challenge also show that there has been considerable progress since the first UERC in 2017, but that there is still ample room for further research in this area.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{grm2019face,

title = {Face Hallucination Revisited: An Exploratory Study on Dataset Bias},

author = {Klemen Grm and Martin Pernus and Leo Cluzel and Walter J. Scheirer and Simon Dobrisek and Vitomir Struc},

url = {http://openaccess.thecvf.com/content_CVPRW_2019/papers/Biometrics/Grm_Face_Hallucination_Revisited_An_Exploratory_Study_on_Dataset_Bias_CVPRW_2019_paper.pdf

https://arxiv.org/pdf/1812.09010.pdf},

year  = {2019},

date = {2019-06-01},

booktitle = {IEEE Conference on Computer Vision and Pattern Recognition Workshops},

abstract = {Contemporary face hallucination (FH) models exhibit considerable ability to reconstruct high-resolution (HR) details from low-resolution (LR) face images.  This ability is commonly learned from examples of corresponding HR-LR image pairs, created by artificially down-sampling the HR ground truth data. This down-sampling (or degradation) procedure not only defines the characteristics of the LR training data, but also determines the type of image degradations the learned FH models are eventually able to handle. If the image characteristics encountered with real-world LR images differ from the ones seen during training, FH models are still expected to perform well, but in practice may not produce the desired results. In this paper we study this problem and explore the bias introduced into FH models by the characteristics of the training data. We systematically analyze the generalization capabilities of several FH models in various scenarios where the degradation function does not match the training setup and conduct experiments with synthetically downgraded as well as real-life low-quality images. We make several interesting findings that provide insight into existing problems with FH models and point to future research directions.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{lozej2019influence,

title = {Influence of segmentation on deep iris recognition performance},

author = {Juš Lozej and Dejan Štepec and Vitomir Štruc and Peter Peer},

url = {https://arxiv.org/pdf/1901.10431.pdf},

year  = {2019},

date = {2019-03-01},

booktitle = {7th IAPR/IEEE International Workshop on Biometrics and Forensics (IWBF 2019)},

journal = {arXiv preprint arXiv:1901.10431},

abstract = {Despite the rise of deep learning in numerous areas of computer vision and image processing, iris recognition has not benefited considerably from these trends so far. Most of the existing research on deep iris recognition is focused on new models for generating discriminative and robust iris representations and relies on methodologies akin to traditional iris recognition pipelines. Hence, the proposed models do not approach iris recognition in an end-to-end manner, but rather use standard heuristic iris segmentation (and unwrapping) techniques to produce normalized inputs for the deep learning models. However, because deep learning is able to model very complex data distributions and nonlinear data changes, an obvious question arises. How important is the use of traditional segmentation methods in a deep learning setting? To answer this question, we present in this paper an empirical analysis of the impact of iris segmentation on the performance of deep learning models using a simple two stage pipeline consisting of a segmentation and a recognition step. We evaluate how the accuracy of segmentation influences recognition performance but also examine if segmentation is needed at all. We use the CASIA Thousand and SBVPI datasets for the experiments and report several interesting findings.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{kristan2018sixth,

title = {The sixth visual object tracking VOT2018 challenge results},

author = {Matej Kristan and Ales Leonardis and Jiri Matas and Michael Felsberg and Roman Pflugfelder and Luka Cehovin Zajc and Tomas Vojir and Goutam Bhat and Alan Lukezic and Abdelrahman Eldesokey and Vitomir Štruc and Klemen Grm and others},

url = {http://openaccess.thecvf.com/content_ECCVW_2018/papers/11129/Kristan_The_sixth_Visual_Object_Tracking_VOT2018_challenge_results_ECCVW_2018_paper.pdf},

year  = {2018},

date = {2018-09-01},

booktitle = {European Conference on Computer Vision Workshops (ECCV-W 2018)},

abstract = {The Visual Object Tracking challenge VOT2018 is the sixth annual tracker benchmarking activity organized by the VOT initiative. Results of over eighty trackers are presented; many are state-of-the-art trackers published at major computer vision conferences or in journals in the recent years. The evaluation included the standard VOT and other popular methodologies for short-term tracking analysis and a “real-time” experiment simulating a situation where a tracker processes images as if provided by a continuously running sensor. A long-term tracking subchallenge has been introduced to the set of standard VOT sub-challenges. The new subchallenge focuses on long-term tracking properties, namely coping with target disappearance and reappearance. A new dataset has been compiled and a performance evaluation methodology that focuses on long-term tracking capabilities has been adopted. The VOT toolkit has been updated to support both standard short-term and the new longterm tracking subchallenges. Performance of the tested trackers typically by far exceeds standard baselines. The source code for most of the trackers is publicly available from the VOT page. The dataset, the evaluation kit and the results are publicly available at the challenge website.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{krivzaj2018localization,

title = {Localization of Facial Landmarks in Depth Images Using Gated Multiple Ridge Descent},

author = {Janez Križaj and Žiga Emeršič and Simon Dobrišek and Peter Peer and Vitomir Štruc},

url = {https://ieeexplore.ieee.org/abstract/document/8464215},

year  = {2018},

date = {2018-09-01},

booktitle = {2018 IEEE International Work Conference on Bioinspired Intelligence (IWOBI)},

pages = {1--8},

organization = {IEEE},

abstract = {A novel method for automatic facial landmark localization is presented. The method builds on the supervised descent framework, which was shown to successfully localize landmarks in the presence of large expression variations and mild occlusions, but struggles when localizing landmarks on faces with large pose variations. We propose an extension of the supervised descent framework that trains multiple descent maps and results in increased robustness to pose variations. The performance of the proposed method is demonstrated on the Bosphorus, the FRGC and the UND data sets for the problem of facial landmark localization from 3D data. Our experimental results show that the proposed method exhibits increased robustness to pose variations, while retaining high performance in the case of expression and occlusion variations.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{lozej2018end,

title = {End-to-end iris segmentation using U-Net},

author = {Juš Lozej and Blaž Meden and Vitomir Struc and Peter Peer},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/IWOBI_2018_paper_15.pdf},

year  = {2018},

date = {2018-07-01},

booktitle = {2018 IEEE International Work Conference on Bioinspired Intelligence (IWOBI)},

pages = {1--6},

organization = {IEEE},

abstract = {Iris segmentation is an important research topic that received significant attention from the research community over the years. Traditional iris segmentation techniques have typically been focused on hand-crafted procedures that, nonetheless, achieved remarkable segmentation performance even with images captured in difficult settings. With the success of deep-learning models, researchers are increasingly looking towards convolutional neural networks (CNNs) to further improve on the accuracy of existing iris segmentation techniques and several CNN-based techniques have already been presented recently in the literature. In this paper we also consider deep-learning models for iris segmentation and present an iris segmentation approach based on the popular U-Net architecture. Our model is trainable end-to-end and, hence, avoids the need for hand designing the segmentation procedure. We evaluate the model on the CASIA dataset and report encouraging results in comparison to existing techniques used in this area.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{rot2018deep,

title = {Deep multi-class eye segmentation for ocular biometrics},

author = {Peter Rot and Žiga Emeršič and Vitomir Struc and Peter Peer},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/MultiClassReduced.pdf},

year  = {2018},

date = {2018-07-01},

booktitle = {2018 IEEE International Work Conference on Bioinspired Intelligence (IWOBI)},

pages = {1--8},

organization = {IEEE},

abstract = {Segmentation techniques for ocular biometrics typically focus on finding a single eye region in the input image at the time. Only limited work has been done on multi-class eye segmentation despite a number of obvious advantages. In this paper we address this gap and present a deep multi-class eye segmentation model build around the SegNet architecture. We train the model on a small dataset (of 120 samples) of eye images and observe it to generalize well to unseen images and to ensure highly accurate segmentation results. We evaluate the model on the Multi-Angle Sclera Database (MASD) dataset and describe comprehensive experiments focusing on: i) segmentation performance, ii) error analysis, iii) the sensitivity of the model to changes in view direction, and iv) comparisons with competing single-class techniques. Our results show that the proposed model is viable solution for multi-class eye segmentation suitable for recognition (multi-biometric) pipelines based on ocular characteristics.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{meden2018selective,

title = {Selective Face Deidentification with End-to-End Perceptual Loss Learning},

author = {Blaz Meden and Peter Peer and Vitomir Struc},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/Selective_Face_Deidentification_with_End_to_End_Perceptual_Loss_Learning.pdf},

year  = {2018},

date = {2018-06-01},

booktitle = {2018 IEEE International Work Conference on Bioinspired Intelligence (IWOBI)},

pages = {1--7},

organization = {IEEE},

abstract = {Privacy is a highly debatable topic in the modern technological era. With the advent of massive video and image data (which in a lot of cases contains personal information on the recorded subjects), there is an imminent need for efficient privacy protection mechanisms. To this end, we develop in this work a novel Face Deidentification Network (FaDeNet) that is able to alter the input faces in such a way that automated recognition fail to recognize the subjects in the images, while this is still possible for human observers. FaDeNet is based an encoder-decoder architecture that is trained to auto-encode the input image, while (at the same time) minimizing the recognition performance of a secondary network that is used as an socalled identity critic in FaDeNet. We present experiments on the Radbound Faces Dataset and observe encouraging results.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{banerjee2018frontalize,

title = {To frontalize or not to frontalize: Do we really need elaborate pre-processing to improve face recognition?},

author = {Sandipan Banerjee and Joel Brogan and Janez Krizaj and Aparna Bharati and Brandon RichardWebster and Vitomir Struc and Patrick J. Flynn and Walter J. Scheirer},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/To_Frontalize_or_Not_To_Frontalize_Do_We_Really_Ne.pdf},

year  = {2018},

date = {2018-05-01},

booktitle = {2018 IEEE Winter Conference on Applications of Computer Vision (WACV)},

pages = {20--29},

organization = {IEEE},

abstract = {Face recognition performance has improved remarkably in the last decade. Much of this success can be attributed to the development of deep learning techniques such as convolutional neural networks (CNNs). While CNNs have pushed the state-of-the-art forward, their training process requires a large amount of clean and correctly labelled training data. If a CNN is intended to tolerate facial pose, then we face an important question: should this training data be diverse in its pose distribution, or should face images be normalized to a single pose in a pre-processing step? To address this question, we evaluate a number of facial landmarking algorithms and a popular frontalization method to understand their effect on facial recognition performance. Additionally, we introduce a new, automatic, single-image frontalization scheme that exceeds the performance of the reference frontalization algorithm for video-to-video face matching on the Point and Shoot Challenge (PaSC) dataset. Additionally, we investigate failure modes of each frontalization method on different facial yaw using the CMU Multi-PIE dataset. We assert that the subsequent recognition and verification performance serves to quantify the effectiveness of each pose correction scheme.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{emervsivc2018towards,

title = {Towards Accessories-Aware Ear Recognition},

author = {Žiga Emeršič and Nil Oleart Playa and Vitomir Štruc and Peter Peer},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/iwobi-2018-inpaint-1.pdf},

doi = {10.1109/IWOBI.2018.8464138},

year  = {2018},

date = {2018-03-01},

booktitle = {2018 IEEE International Work Conference on Bioinspired Intelligence (IWOBI)},

pages = {1--8},

organization = {IEEE},

abstract = {Automatic ear recognition is gaining popularity within the research community due to numerous desirable properties, such as high recognition performance, the possibility of capturing ear images at a distance and in a covert manner, etc. Despite this popularity and the corresponding research effort that is being directed towards ear recognition technology, open problems still remain. One of the most important issues stopping ear recognition systems from being widely available are ear occlusions and accessories. Ear accessories not only mask biometric features and by this reduce the overall recognition performance, but also introduce new non-biometric features that can be exploited for spoofing purposes. Ignoring ear accessories during recognition can, therefore, present a security threat to ear recognition and also adversely affect performance. Despite the importance of this topic there has been, to the best of our knowledge, no ear recognition studies that would address these problems. In this work we try to close this gap and study the impact of ear accessories on the recognition performance of several state-of-the-art ear recognition techniques. We consider ear accessories as a tool for spoofing attacks and show that CNN-based recognition approaches are more susceptible to spoofing attacks than traditional descriptor-based approaches. Furthermore, we demonstrate that using inpainting techniques or average coloring can mitigate the problems caused by ear accessories and slightly outperforms (standard) black color to mask ear accessories.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Dasicb2018,

title = {SSBC 2018: Sclera Segmentation Benchmarking Competition},

author = {Abhijit Das and Umapada Pal and Miguel A. Ferrer and Michael Blumenstein and Dejan Štepec and Peter Rot and Žiga Emeršič and Peter Peer and Vitomir Štruc},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/icb2018_sserbc.pdf},

year  = {2018},

date = {2018-02-01},

booktitle = {2018 International Conference on Biometrics (ICB)},

abstract = {This paper summarises the results of the Sclera Segmentation Benchmarking Competition (SSBC 2018). It was organised in the context of the 11th IAPR International Conference on Biometrics (ICB 2018). The aim of this competition was to record the developments on sclera segmentation in the cross-sensor environment (sclera trait captured using multiple acquiring sensors). Additionally, the competition also aimed to gain the attention of researchers on this subject of research. For the purpose of benchmarking, we have developed two datasets of sclera images captured using different sensors. The first dataset was collected using a DSLR camera and the second one was collected using a mobile phone camera. The first dataset is the Multi-Angle Sclera Dataset (MASD version 1), which was used in the context of the previous versions of sclera segmentation competitions. The images in the second dataset were captured using .a mobile phone rear camera of 8-megapixel. As baseline manual segmentation mask of the sclera images from both the datasets were developed. Precision and recall-based statistical measures were employed to evaluate the effectiveness of the submitted segmentation technique and to rank them. Six algorithms were submitted towards the segmentation task. This paper analyses the results produced by these algorithms/system and defines a way forward for this subject of research. Both the datasets along with some of the accompanying ground truth/baseline mask will be freely available for research purposes upon request to authors by email.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{vidal2018ug,

title = {UG^ 2: A Video Benchmark for Assessing the Impact of Image Restoration and Enhancement on Automatic Visual Recognition},

author = {Rosaura G. Vidal and Sreya Banerjee and Klemen Grm and Vitomir Struc and Walter J. Scheirer},

url = {https://arxiv.org/pdf/1710.02909.pdf},

year  = {2018},

date = {2018-02-01},

booktitle = {2018 IEEE Winter Conference on Applications of Computer Vision (WACV)},

pages = {1597--1606},

organization = {IEEE},

abstract = {Advances in image restoration and enhancement techniques have led to discussion about how such algorithms can be applied as a pre-processing step to improve automatic visual recognition. In principle, techniques like deblurring and super-resolution should yield improvements by de-emphasizing noise and increasing signal in an input image. But the historically divergent goals of computational photography and visual recognition communities have created a significant need for more work in this direction. To facilitate new research, we introduce a new benchmark dataset called UG2, which contains three difficult real-world scenarios: uncontrolled videos taken by UAVs and manned gliders, as well as controlled videos taken on the ground. Over 150,000 annotated frames for hundreds of ImageNet classes are available, which are used for baseline experiments that assess the impact of known and unknown image artifacts and other conditions on common deep learning-based object classification approaches. Further, current image restoration and enhancement techniques are evaluated by determining whether or not they improve baseline classification performance. Results show that there is plenty of room for algorithmic innovation, making this dataset a useful tool going forward.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{ERK2017,

title = {Do it Yourself: Building a Low-Cost Iris Recognition System at Home Using Off-The-Shelf Components},

author = {Primož Lavrič and Žiga Emeršič and Blaž Meden and Vitomir Štruc and Peter Peer},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/lavricdo_it.pdf},

year  = {2017},

date = {2017-09-01},

booktitle = {Electrotechnical and Computer Science Conference ERK 2017},

abstract = {Among the different biometric traits that can be used for person recognition,  the human iris is generally consid-ered to be among the most accurate.   However,  despite a plethora of desirable characteristics, iris recognition is not widely as widely used as competing biometric modalities likely due to the high cost of existing commercial iris-recognition systems. In this paper we contribute towards the availability of low-cost iris recognition systems and present a prototype system built using off-the-shelf components.  We describe the prototype device, the pipeline used for iris recognition, evaluate the performance of our solution on a small in-house dataset and discuss directions for future work.  The current version of our prototype includes complete hardware and software implementations and has a combined bill-of-materials of 110 EUR. 



},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{emervsivc2017unconstrained,

title = {The unconstrained ear recognition challenge},

author = {Žiga Emeršič and Dejan Štepec and Vitomir Štruc and Peter Peer and Anjith George and Adii Ahmad and Elshibani Omar and Terrance E. Boult and Reza Safdaii and Yuxiang Zhou and others Stefanos Zafeiriou and Dogucan Yaman and Fevziye I. Eyoikur and Hazim K. Ekenel},

url = {https://arxiv.org/pdf/1708.06997.pdf},

year  = {2017},

date = {2017-09-01},

booktitle = {2017 IEEE International Joint Conference on Biometrics (IJCB)},

pages = {715--724},

organization = {IEEE},

abstract = {In this paper we present the results o f the Unconstrained Ear Recognition Challenge (UERC), a group benchmarking effort centered around the problem o f person recognition from ear images captured in uncontrolled conditions. The goal o f the challenge was to assess the performance of existing ear recognition techniques on a challenging largescale dataset and identify open problems that need to be addressed in the future. Five groups from three continents participated in the challenge and contributed six ear recognition techniques fo r the evaluation, while multiple baselines were made available for the challenge by the UERC organizers. A comprehensive analysis was conducted with all participating approaches addressing essential research questions pertaining to the sensitivity o f the technology to head rotation, flipping, gallery size, large-scale recognition and others. The top performer o f the UERC was found to ensure robust performance on a smaller part o f the dataset (with 180 subjects) regardless o f image characteristics, but still exhibited a significant performance drop when the entire dataset comprising 3,704 subjects was used for testing.

},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{ERK2017Grm,

title = {Evaluating image superresolution algorithms for cross-resolution face recognition},

author = {Grm Klemen and Dobrišek Simon and Štruc Vitomir},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/review_submission.pdf},

year  = {2017},

date = {2017-09-01},

booktitle = {Proceedings of the Twenty-sixth International Electrotechnical and Computer Science Conference ERK 2017},

abstract = {With recent advancements in deep learning and convolutional neural networks (CNNs), face recognition has seen significant performance improvements over the last few years. However, low-resolution images still remain challenging, with CNNs performing relatively poorly compared to humans. One possibility to improve performance in these settings often advocated in the literature is the use of super-resolution (SR). In this paper, we explore the usefulness of SR algorithms for cross-resolution face recognition in experiments on the Labeled Faces in the Wild (LFW) and SCface datasets using four recent deep CNN models. We conduct experiments with synthetically down-sampled images as well as real-life low-resolution imagery captured by surveillance cameras. Our experiments show that image super-resolution can improve face recognition performance considerably on very low-resolution images (of size 24 x 24 or 32 x 32 pixels), when images are artificially down-sampled, but has a lesser (or sometimes even a detrimental) effect with real-life images leaving significant room for further research in this area.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{ERK2017c,

title = {Face recognition with Raspberry Pi for IoT Environments.},

author = {Novosel Rok and Meden Blaž and Emeršič Žiga and Štruc Vitomir and Peter Peer},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/novoselface_recognition.pdf},

year  = {2017},

date = {2017-09-01},

booktitle = {Proceedings of the Twenty-sixth International Electrotechnical and Computer Science Conference ERK 2017},

abstract = {IoT has seen steady growth over recent years – smart home appliances, smart personal gear, personal assistants and many more. The same is true for the field of bio-metrics where the need for automatic and secure recognition schemes have spurred the development of fingerprint-and face-recognition mechanisms found today in most smart phones and similar hand-held devices. Devices used in the Internet of Things (IoT) are often low-powered with limited computational resources. This means that biomet-ric recognition pipelines aimed at IoT need to be streamlined and as efficient as possible. Towards this end, we describe in this paper how image-based biometrics can be leveraged in an IoT environment using a Raspberry Pi. We present a proof-of-concept web-based information system, secured by a face-recognition procedure, that gives authorized users access to potentially sensitive information.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{emervsivc2017training,

title = {Training convolutional neural networks with limited training data for ear recognition in the wild},

author = {Žiga Emeršič and Dejan Štepec and Vitomir Štruc and Peter Peer},

url = {https://arxiv.org/pdf/1711.09952.pdf},

year  = {2017},

date = {2017-05-01},

booktitle = {IEEE International Conference on Automatic Face and Gesture Recognition, Workshop on Biometrics in the Wild 2017},

journal = {arXiv preprint arXiv:1711.09952},

abstract = {Identity recognition from ear images is an active field of research within the biometric community. The ability to capture ear images from a distance and in a covert manner makes ear recognition technology an appealing choice for surveillance and security applications as well as related application domains. In contrast to other biometric modalities, where large datasets captured in uncontrolled settings are readily available, datasets of ear images are still limited in size and mostly of laboratory-like quality. As a consequence, ear recognition technology has not benefited yet from advances in deep learning and convolutional neural networks (CNNs) and is still lacking behind other modalities that experienced significant performance gains owing to deep recognition technology. In this paper we address this problem and aim at building a CNNbased ear recognition model. We explore different strategies towards model training with limited amounts of training data and show that by selecting an appropriate model architecture, using aggressive data augmentation and selective learning on existing (pre-trained) models, we are able to learn an effective CNN-based model using a little more than 1300 training images. The result of our work is the first CNN-based approach to ear recognition that is also made publicly available to the research community. With our model we are able to improve on the rank one recognition rate of the previous state-of-the-art by more than 25% on a challenging dataset of ear images captured from the web (a.k.a. in the wild).},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{emersic2017covariate,

title = {Covariate analysis of descriptor-based ear recognition techniques},

author = {Ziga Emersic and Blaz Meden and Peter Peer and Vitornir Struc},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/Covariate_Analysis_of_Descriptor_based_Ear_Recognition_Techniques.pdf},

year  = {2017},

date = {2017-01-01},

booktitle = {2017 international conference and workshop on bioinspired intelligence (IWOBI)},

pages = {1--9},

organization = {IEEE},

abstract = {Dense descriptor-based feature extraction techniques represent a popular choice for implementing biometric ear recognition system and are in general considered to be the current state-of-the-art in this area. In this paper, we study the impact of various factors (i.e., head rotation, presence of occlusions, gender and ethnicity) on the performance of 8 state-of-the-art descriptor-based ear recognition techniques. Our goal is to pinpoint weak points of the existing technology and identify open problems worth exploring in the future. We conduct our covariate analysis through identification experiments on the challenging AWE (Annotated Web Ears) dataset and report our findings. The results of our study show that high degrees of head movement and presence of accessories significantly impact the identification performance, whereas mild degrees of the listed factors and other covariates such as gender and ethnicity impact the identification performance only to a limited extent.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{meden2017kappa,

title = {k-Same-Net: Neural-Network-Based Face Deidentification},

author = {Blaz Meden and Ziga Emersic and Vitomir Struc and Peter Peer},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/k-same-net.pdf},

year  = {2017},

date = {2017-01-01},

booktitle = {2017 International Conference and Workshop on Bioinspired Intelligence (IWOBI)},

pages = {1--7},

organization = {IEEE},

abstract = {An increasing amount of video and image data is being shared between government entities and other relevant stakeholders and requires careful handling of personal information. A popular approach for privacy protection in such data is the use of deidentification techniques, which aim at concealing the identity of individuals in the imagery while still preserving certain aspects of the data deidentification. In this work, we propose a novel approach towards face deidentification, called k-Same-Net, which combines recent generative neural networks (GNNs) with the well-known k-anonymity mechanism and provides formal guarantees regarding privacy protection on a closed set of identities. Our GNN is able to generate synthetic surrogate face images for dedentification by seamlessly combining features of identities used to train the GNN mode. furthermore, it allows us to guide the image-generation process with a small set of appearance-related parameters that can be used to alter specific aspects (e.g., facial expressions, age, gender) of the synthesized surrogate images. We demonstrate the feasibility of k-Same-Net in comparative experiments with competing techniques on the XM2VTS dataset and discuss the main characteristics of our approach.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{das2017sserbc,

title = {SSERBC 2017: Sclera segmentation and eye recognition benchmarking competition},

author = {Abhijit Das and Umapada Pal and Miguel A Ferrer and Michael Blumenstein and Dejan Štepec and Peter Rot and Ziga Emeršič and Peter Peer and Vitomir Štruc and SV Aruna Kumar and Harish B S},

url = {https://lmi.fe.uni-lj.si/wp-content/uploads/2019/08/SSERBC2017.pdf},

year  = {2017},

date = {2017-01-01},

booktitle = {2017 IEEE International Joint Conference on Biometrics (IJCB)},

pages = {742--747},

organization = {IEEE},

abstract = {This paper summarises the results of the Sclera Segmentation and Eye Recognition Benchmarking Competition (SSERBC 2017). It was organised in the context of the International Joint Conference on Biometrics (IJCB 2017). The aim of this competition was to record the recent developments in sclera segmentation and eye recognition in the visible spectrum (using iris, sclera and peri-ocular, and their fusion), and also to gain the attention of researchers on this subject.



In this regard, we have used the Multi-Angle Sclera Dataset (MASD version 1). It is comprised of 2624 images taken from both the eyes of 82 identities. Therefore, it consists of images of 164 (82*2) eyes. A manual segmentation mask of these images was created to baseline both tasks.



Precision and recall based statistical measures were employed to evaluate the effectiveness of the segmentation and the ranks of the segmentation task. Recognition accuracy measure has been employed to measure the recognition task. Manually segmented sclera, iris and periocular regions were used in the recognition task. Sixteen teams registered for the competition, and among them, six teams submitted their algorithms or systems for the segmentation task and two of them submitted their recognition algorithm or systems.



The results produced by these algorithms or systems reflect current developments in the literature of sclera segmentation and eye recognition, employing cutting edge techniques. The MASD version 1 dataset with some of the ground truth will be freely available for research purposes. The success of the competition also demonstrates the recent interests of researchers from academia as well as industry on this subject},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{BTAS2016,

title = {Report on the BTAS 2016 Video Person Recognition Evaluation},

author = {Walter Scheirer and Patrick Flynn and Changxing Ding and Guodong Guo and Vitomir Štruc and Mohamad Al Jazaery and Simon Dobrišek and Klemen Grm and Dacheng Tao and Yu Zhu and Joel Brogan and Sandipan Banerjee and Aparna Bharati and Brandon Richard Webster},

year  = {2016},

date = {2016-10-05},

booktitle = {Proceedings of the IEEE International Conference on Biometrics: Theory, Applications ans Systems (BTAS)},

publisher = {IEEE},

abstract = {This report presents results from the Video Person Recognition Evaluation held in conjunction with the 8th IEEE International Conference on Biometrics: Theory, Applications, and Systems (BTAS). Two experiments required algorithms to recognize people in videos from the Pointand- Shoot Face Recognition Challenge Problem (PaSC). The first consisted of videos from a tripod mounted high quality video camera. The second contained videos acquired from 5 different handheld video cameras. There were 1,401 videos in each experiment of 265 subjects. The subjects, the scenes, and the actions carried out by the people are the same in both experiments. An additional experiment required algorithms to recognize people in videos from the Video Database of Moving Faces and People (VDMFP). There were 958 videos in this experiment of 297 subjects. Four groups from around the world participated in the evaluation. The top verification rate for PaSC from this evaluation is 0:98 at a false accept rate of 0:01 — a remarkable advancement in performance from the competition held at FG 2015.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{ERK2016Janez,

title = {Facial Landmark Localization from 3D Images},

author = {Janez Križaj and Simon Dobrišek and France Mihelič and Vitomir Štruc},

year  = {2016},

date = {2016-09-20},

booktitle = {Proceedings of the Electrotechnical and Computer Science Conference (ERK)},

address = {Portorož, Slovenia},

abstract = {A novel method for automatic facial landmark localization is presented. The method builds on the supervised descent framework, which was shown to successfully localize landmarks in the presence of large expression variations and mild occlusions, but struggles when localizing landmarks on faces with large pose variations. We propose an extension of the supervised descent framework which trains multiple descent maps and results in increased robustness to pose variations. The performance of the proposed method is demonstrated on the Bosphorus database for the problem of facial landmark localization from 3D data. Our experimental results show that the proposed method exhibits increased robustness to pose variations, while retaining high performance in the case of expression and occlusion variations.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{ERK2016_Seba,

title = {Vpliv registracije obraznih področij na učinkovitost samodejnega razpoznavanja obrazov: študija z OpenBR},

author = {Sebastjan Fabijan and Vitomir Štruc},

url = {https://lmi.fe.uni-lj.si/en/vplivregistracijeobraznihpodrocijnaucinkovitostsamodejnegarazpoznavanjaobrazovstudijazopenbr/},

year  = {2016},

date = {2016-09-20},

urldate = {2016-09-20},

booktitle = {Proceedings of the Electrotechnical and Computer Science Conference (ERK)},

abstract = {Razpoznavanje obrazov je v zadnjih letih postalo eno najuspešnejših področij samodejne, računalniško podprte analize slik, ki se lahko pohvali z različnimi primeri upor-abe v praksi. Enega ključnih korakav za uspešno razpoznavanje predstavlja poravnava obrazov na slikah. S poravnavo poskušamo zagotoviti neodvisnost razpozn-av-an-ja od sprememb zornih kotov pri zajemu slike, ki v slikovne podatke vnašajo visoko stopnjo variabilnosti. V prispevku predstavimo tri postopke poravnavanja obrazov (iz literature) in proučimo njihov vpliv na uspešnost razpoznavanja s postopki, udejanjenimi v odprtokodnem programskem ogrodju Open Source Biometric Recognition (OpenBR). Vse poizkuse izvedemo na podatkovni zbirki Labeled Faces in the Wild (LFW).},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{ERK2016_sebastjan,

title = {Učenje podobnosti v globokih nevronskih omrežjih za razpoznavanje obrazov},

author = {Žiga Stržinar and Klemen Grm and Vitomir Štruc},

url = {https://lmi.fe.uni-lj.si/en/ucenjepodobnostivglobokihnevronskihomrezjihzarazpoznavanjeobrazov/},

year  = {2016},

date = {2016-09-20},

urldate = {2016-09-20},

booktitle = {Proceedings of the Electrotechnical and Computer Science Conference (ERK)},

address = {Portorož, Slovenia},

abstract = {Učenje podobnosti med pari vhodnih slik predstavlja enega najpopularnejših pristopov k razpoznavanju na področju globokega učenja. Pri tem pristopu globoko nevronsko omrežje na vhodu sprejme par slik (obrazov) in na izhodu vrne mero podobnosti med vhodnima slikama, ki jo je moč uporabiti za razpoznavanje. Izračun podobnosti je pri tem lahko v celoti udejanjen z globokim omrežjem, lahko pa se omrežje uporabi zgolj za izračun predstavitve vhodnega para slik, preslikava iz izračunane predstavitve v mero podobnosti pa se izvede z drugim, potencialno primernejšim modelom. V tem prispevku preizkusimo 5 različnih modelov za izvedbo preslikave med izračunano predstavitvijo in mero podobnosti, pri čemer za poizkuse uporabimo lastno nevronsko omrežje. Rezultati naših eksperimentov na problemu razpoznavanja obrazov kažejo na pomembnost izbire primernega modela, saj so razlike med uspešnostjo razpoznavanje od modela do modela precejšnje.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{SJDT,

title = {Assessment of the Google Speech Application Programming Interface for Automatic Slovenian Speech Recognition},

author = {Simon Dobrišek and David Čefarin and Vitomir Štruc and France Mihelič},

url = {https://lmi.fe.uni-lj.si/en/assessmentofthegooglespeechapplicationprogramminginterfaceforautomaticslovenianspeechrecognition/},

year  = {2016},

date = {2016-09-20},

urldate = {2016-09-20},

booktitle = {Jezikovne Tehnologije in Digitalna Humanistika},

abstract = {Automatic speech recognizers are slowly maturing into technologies that enable humans to communicate more naturally and effectively with a variety of smart devices and information-communication systems. Large global companies such as Google, Microsoft, Apple, IBM and Baidu compete in developing the most reliable speech recognizers, supporting as many of the main world languages as possible. Due to the relatively small number of speakers, the support for the Slovenian spoken language is lagging behind, and among the major global companies only Google has recently supported our spoken language. The paper presents the results of our independent assessment of the Google speech-application programming interface for automatic Slovenian speech recognition. For the experiments, we used speech databases that are otherwise used for the development and assessment of Slovenian speech recognizers.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{RibicERK2016,

title = {Influence of alignment on ear recognition : case study on AWE Dataset},

author = {Metod Ribič and Žiga Emeršič and Vitomir Štruc and Peter Peer},

url = {https://lmi.fe.uni-lj.si/en/influenceofalignmentonearrecognitioncasestudyonawedataset/},

year  = {2016},

date = {2016-09-20},

urldate = {2016-09-20},

booktitle = {Proceedings of the Electrotechnical and Computer Science Conference (ERK)},

pages = {131-134},

address = {Portorož, Slovenia},

abstract = {Ear as a biometric modality presents a viable source for automatic human recognition. In recent years local description methods have been gaining on popularity due to their invariance to illumination and occlusion. However, these methods require that images are well aligned and preprocessed as good as possible. This causes one of the greatest challenges of ear recognition: sensitivity to pose variations. Recently, we presented Annotated Web Ears dataset that opens new challenges in ear recognition. In this paper we test the influence of alignment on recognition performance and prove that even with the alignment the database is still very challenging, even-though the recognition rate is improved due to alignment. We also prove that more sophisticated alignment methods are needed to address the AWE dataset efficiently},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{dobrivsekpreizkus,

title = {Preizkus Googlovega govornega programskega vmesnika pri samodejnem razpoznavanju govorjene slovenščine},

author = {Simon Dobrišek and David Čefarin and Vitomir Štruc and France Mihelič},

url = {http://www.sdjt.si/wp/wp-content/uploads/2016/09/JTDH-2016_Dobrisek-et-al_Preizkus-Googlovega-govornega-programskega-vmesnika.pdf},

year  = {2016},

date = {2016-09-01},

booktitle = {Jezikovne tehnologije in digitalna humanistika},

pages = {47-51},

abstract = {Automatic speech recognizers are slowly maturing into technologies that enable humans to communicate more naturally and effectively with a variety of smart devices and information-communication systems. Large global companies such as Google, Microsoft, Apple, IBM and Baidu compete in developing the most reliable speech recognizers, supporting as many of the main world languages as possible. Due to the relatively small number of speakers, the support for the Slovenian spoken language is lagging behind, and among the major global companies only Google has recently supported our spoken language. The paper presents the results of our independent assessment of the Google speech-application programming interface for automatic Slovenian speech recognition. For the experiments, we used speech databases that are otherwise used for the development and assessment of Slovenian speech recognizers. 

},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{golob2016composition,

title = {A Composition Algorithm of Compact Finite-State Super Transducers for Grapheme-to-Phoneme Conversion},

author = {Žiga Golob and Jerneja Žganec Gros and Vitomir Štruc and France Mihelič and Simon Dobrišek},

year  = {2016},

date = {2016-01-01},

booktitle = {International Conference on Text, Speech, and Dialogue},

pages = {375--382},

organization = {Springer},

abstract = {Minimal deterministic finite-state transducers (MDFSTs) are powerful models that can be used to represent pronunciation dictionaries in a compact form. Intuitively, we would assume that by increasing the size of the dictionary, the size of the MDFSTs would increase as well. However, as we show in the paper, this intuition does not hold for highly inflected languages. With such languages the size of the MDFSTs begins to decrease once the number of words in the represented dictionary reaches a certain threshold. Motivated by this observation, we have developed a new type of FST, called a finite-state super transducer (FSST), and show experimentally that the FSST is capable of representing pronunciation dictionaries with fewer states and transitions than MDFSTs. Furthermore, we show that (unlike MDFSTs) our FSSTs can also accept words that are not part of the represented dictionary. The phonetic transcriptions of these out-of-dictionary words may not always be correct, but the observed error rates are comparable to the error rates of the traditional methods for grapheme-to-phoneme conversion.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{grm2016deep,

title = {Deep pair-wise similarity learning for face recognition},

author = {Klemen Grm and Simon Dobrišek and Vitomir Štruc},

url = {https://lmi.fe.uni-lj.si/en/deeppair-wisesimilaritylearningforfacerecognition/},

year  = {2016},

date = {2016-01-01},

urldate = {2016-01-01},

booktitle = {4th International Workshop on Biometrics and Forensics (IWBF)},

pages = {1--6},

organization = {IEEE},

abstract = {Recent advances in deep learning made it possible to build deep hierarchical models capable of delivering state-of-the-art performance in various vision tasks, such as object recognition, detection or tracking. For recognition tasks the most common approach when using deep models is to learn object representations (or features) directly from raw image-input and then feed the learned features to a suitable classifier. Deep models used in this pipeline are typically heavily parameterized and require enormous amounts of training data to deliver competitive recognition performance. Despite the use of data augmentation techniques, many application domains, predefined experimental protocols or specifics of the recognition problem limit the amount of available training data and make training an effective deep hierarchical model a difficult task. In this paper, we present a novel, deep pair-wise similarity learning (DPSL) strategy for deep models, developed specifically to overcome the problem of insufficient training data, and demonstrate its usage on the task of face recognition. Unlike existing (deep) learning strategies, DPSL operates on image-pairs and tries to learn pair-wise image similarities that can be used for recognition purposes directly instead of feature representations that need to be fed to appropriate classification techniques, as with traditional deep learning pipelines. Since our DPSL strategy assumes an image pair as the input to the learning procedure, the amount of training data available to train deep models is quadratic in the number of available training images, which is of paramount importance for models with a large number of parameters. We demonstrate the efficacy of the proposed learning strategy by developing a deep model for pose-invariant face recognition, called Pose-Invariant Similarity Index (PISI), and presenting comparative experimental results on the FERET an IJB-A datasets.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{ERK2015Klemen,

title = {The pose-invariant similarity index for face recognition},

author = {Klemen Grm and Simon Dobrišek and Vitomir Štruc},

year  = {2015},

date = {2015-04-20},

booktitle = {Proceedings of the Electrotechnical and Computer Science Conference (ERK)},

address = {Portorož, Slovenia},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{struc2015modest,

title = {Modest face recognition},

author = {Vitomir Štruc and Janez Križaj and Simon Dobrišek},

url = {https://lmi.fe.uni-lj.si/en/modestfacerecognition/},

year  = {2015},

date = {2015-01-01},

urldate = {2015-01-01},

booktitle = {Proceedings of the International Workshop on Biometrics and Forensics (IWBF)},

pages = {1--6},

publisher = {IEEE},

abstract = {The facial imagery usually at the disposal for forensics investigations is commonly of a poor quality due to the unconstrained settings in which it was acquired. The captured faces are typically non-frontal, partially occluded and of a low resolution, which makes the recognition task extremely difficult. In this paper we try to address this problem by presenting a novel framework for face recognition that combines diverse features sets (Gabor features, local binary patterns, local phase quantization features and pixel intensities), probabilistic linear discriminant analysis (PLDA) and data fusion based on linear logistic regression. With the proposed framework a matching score for the given pair of probe and target images is produced by applying PLDA on each of the four feature sets independently - producing a (partial) matching score for each of the PLDA-based feature vectors - and then combining the partial matching results at the score level to generate a single matching score for recognition. We make two main contributions in the paper: i) we introduce a novel framework for face recognition that relies on probabilistic MOdels of Diverse fEature SeTs (MODEST) to facilitate the recognition process and ii) benchmark it against the existing state-of-the-art. We demonstrate the feasibility of our MODEST framework on the FRGCv2 and PaSC databases and present comparative results with the state-of-the-art recognition techniques, which demonstrate the efficacy of our framework.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{beveridge2015report,

title = {Report on the FG 2015 video person recognition evaluation},

author = {Ross Beveridge and Hao Zhang and Bruce A Draper and Patrick J Flynn and Zhenhua Feng and Patrik Huber and Josef Kittler and Zhiwu Huang and Shaoxin Li and Yan Li and Vitomir Štruc and Janez Križaj and others},

url = {https://lmi.fe.uni-lj.si/en/reportonthefg2015videopersonrecognitionevaluation/},

year  = {2015},

date = {2015-01-01},

urldate = {2015-01-01},

booktitle = {11th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition (IEEE FG)},

volume = {1},

pages = {1--8},

organization = {IEEE},

abstract = {This report presents results from the Video Person Recognition Evaluation held in conjunction with the 11th IEEE International Conference on Automatic Face and Gesture Recognition. Two experiments required algorithms to recognize people in videos from the Point-and-Shoot Face Recognition Challenge Problem (PaSC). The first consisted of videos from a tripod mounted high quality video camera. The second contained videos acquired from 5 different handheld video cameras. There were 1401 videos in each experiment of 265 subjects. The subjects, the scenes, and the actions carried out by the people are the same in both experiments. Five groups from around the world participated in the evaluation. The video handheld experiment was included in the International Joint Conference on Biometrics (IJCB) 2014 Handheld Video Face and Person Recognition Competition. The top verification rate from this evaluation is double that of the top performer in the IJCB competition. Analysis shows that the factor most effecting algorithm performance is the combination of location and action: where the video was acquired and what the person was doing.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{justin2015speaker,

title = {Speaker de-identification using diphone recognition and speech synthesis},

author = {Tadej Justin and Vitomir Štruc and Simon Dobrišek and Boštjan Vesnicer and Ivo Ipšić and France Mihelič},

url = {https://lmi.fe.uni-lj.si/en/speakerde-identificationusingdiphonerecognitionandspeechsynthesis/},

year  = {2015},

date = {2015-01-01},

urldate = {2015-01-01},

booktitle = {11th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition (IEEE FG): DeID 2015},

volume = {4},

pages = {1--7},

organization = {IEEE},

abstract = {The paper addresses the problem of speaker (or voice) de-identification by presenting a novel approach for concealing the identity of speakers in their speech. The proposed technique first recognizes the input speech with a diphone recognition system and then transforms the obtained phonetic transcription into the speech of another speaker with a speech synthesis system. Due to the fact that a Diphone RecOgnition step and a sPeech SYnthesis step are used during the deidentification, we refer to the developed technique as DROPSY. With this approach the acoustical models of the recognition and synthesis modules are completely independent from each other, which ensures the highest level of input speaker deidentification. The proposed DROPSY-based de-identification approach is language dependent, text independent and capable of running in real-time due to the relatively simple computing methods used. When designing speaker de-identification technology two requirements are typically imposed on the deidentification techniques: i) it should not be possible to establish the identity of the speakers based on the de-identified speech, and ii) the processed speech should still sound natural and be intelligible. This paper, therefore, implements the proposed DROPSY-based approach with two different speech synthesis techniques (i.e, with the HMM-based and the diphone TDPSOLA- based technique). The obtained de-identified speech is evaluated for intelligibility and evaluated in speaker verification experiments with a state-of-the-art (i-vector/PLDA) speaker recognition system. The comparison of both speech synthesis modules integrated in the proposed method reveals that both can efficiently de-identify the input speakers while still producing intelligible speech.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{dobrivsek2015face,

title = {Face recognition in the wild with the Probabilistic Gabor-Fisher Classifier},

author = {Simon Dobrišek and Vitomir Štruc and Janez Križaj and France Mihelič},

url = {https://lmi.fe.uni-lj.si/en/facerecognitioninthewildwiththeprobabilisticgabor-fisherclassifier/},

year  = {2015},

date = {2015-01-01},

urldate = {2015-01-01},

booktitle = {11th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition (IEEE FG): BWild 2015},

volume = {2},

pages = {1--6},

organization = {IEEE},

abstract = {The paper addresses the problem of face recognition in the wild. It introduces a novel approach to unconstrained face recognition that exploits Gabor magnitude features and a simplified version of the probabilistic linear discriminant analysis (PLDA). The novel approach, named Probabilistic Gabor-Fisher Classifier (PGFC), first extracts a vector of Gabor magnitude features from the given input image using a battery of Gabor filters, then reduces the dimensionality of the extracted feature vector by projecting it into a low-dimensional subspace and finally produces a representation suitable for identity inference by applying PLDA to the projected feature vector. The proposed approach extends the popular Gabor-Fisher Classifier (GFC) to a probabilistic setting and thus improves on the generalization capabilities of the GFC method. The PGFC technique is assessed in face verification experiments on the Point and Shoot Face Recognition Challenge (PaSC) database, which features real-world videos of subjects performing everyday tasks. Experimental results on this challenging database show the feasibility of the proposed approach, which improves on the best results on this database reported in the literature by the time of writing.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{justin2015development,

title = {Development and Evaluation of the Emotional Slovenian Speech Database-EmoLUKS},

author = {Tadej Justin and Vitomir Štruc and Janez Žibert and France Mihelič},

url = {https://lmi.fe.uni-lj.si/en/developmentandevaluationoftheemotionalslovenianspeechdatabase-emoluks/},

year  = {2015},

date = {2015-01-01},

urldate = {2015-01-01},

booktitle = {Proceedings of the International Conference on Text, Speech, and Dialogue (TSD)},

pages = {351--359},

organization = {Springer},

abstract = {This paper describes a speech database built from 17 Slovenian radio dramas. The dramas were obtained from the national radio-and-television station (RTV Slovenia) and were given at the universities disposal with an academic license for processing and annotating the audio material. The utterances of one male and one female speaker were transcribed, segmented and then annotated with emotional states of the speakers. The annotation of the emotional states was conducted in two stages with our own web-based application for crowd sourcing. The final (emotional) speech database consists of 1385 recordings of one male (975 recordings) and one female (410 recordings) speaker and contains labeled emotional speech with a total duration of around 1 hour and 15 minutes. The paper presents the two-stage annotation process used to label the data and demonstrates the usefulness of the employed annotation methodology. Baseline emotion recognition experiments are also presented. The reported results are presented with the un-weighted as well as weighted average recalls and precisions for 2-class and 7-class recognition experiments.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{cihan2015facial,

title = {Facial Landmark Localization in Depth Images using Supervised Ridge Descent},

author = {Necati Cihan Camgoz and Vitomir Štruc and Berk Gokberk and Lale Akarun and Ahmet Alp Kindiroglu},

url = {https://lmi.fe.uni-lj.si/en/faciallandmarklocalizationindepthimagesusingsupervisedridgedescent/},

year  = {2015},

date = {2015-01-01},

urldate = {2015-01-01},

booktitle = {Proceedings of the IEEE International Conference on Computer Vision Workshops (ICCVW): Chaa Learn},

pages = {136--141},

abstract = {Supervised Descent Method (SDM) has proven successful in many computer vision applications such as face alignment, tracking and camera calibration. Recent studies which used SDM, achieved state of the-art performance on facial landmark localization in depth images [4]. In this study, we propose to use ridge regression instead of least squares regression for learning the SDM, and to change feature sizes in each iteration, effectively turning the landmark search into a coarse to fine process. We apply the proposed method to facial landmark localization on the Bosphorus 3D Face Database; using frontal depth images with no occlusion. Experimental results confirm that both ridge regression and using adaptive feature sizes improve the localization accuracy considerably},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{krivzaj2014feasibility,

title = {A Feasibility Study on the Use of Binary Keypoint Descriptors for 3D Face Recognition},

author = {Janez Križaj and Vitomir Štruc and France Mihelič},

url = {https://lmi.fe.uni-lj.si/en/afeasibilitystudyontheuseofbinarykeypointdescriptorsfor3dfacerecognition/},

year  = {2014},

date = {2014-01-01},

urldate = {2014-01-01},

booktitle = {Proceedings of the Mexican Conference on Pattern Recognition (MCPR)},

pages = {142--151},

organization = {Springer},

abstract = {Despite the progress made in the area of local image descriptors in recent years, virtually no literature is available on the use of more recent descriptors for the problem of 3D face recognition, such as BRIEF, ORB, BRISK or FREAK, which are binary in nature and, therefore, tend to be faster to compute and match, while requiring signicantly less memory for storage than, for example, SIFT or SURF. In this paper, we try to close this gap and present a feasibility study on the use of these descriptors for 3D face recognition. Descriptors are evaluated on the three challenging 3D face image datasets, namely, the FRGC, UMB and CASIA. Our experiments show the binary descriptors ensure slightly lower verication rates than SIFT, comparable to those of the SURF descriptor, while being an order of magnitude faster than SIFT. The results suggest that the use of binary descriptors represents a viable alternative to the established descriptors.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{krivzaj2014sift,

title = {SIFT vs. FREAK: Assessing the usefulness of two keypoint descriptors for 3D face verification},

author = {Janez Križaj and Vitomir Štruc and Simon Dobrišek and Darijan Marčetić and Slobodan Ribarić},

url = {https://lmi.fe.uni-lj.si/en/siftvs-freakassessingtheusefulnessoftwokeypointdescriptorsfor3dfaceverification/},

year  = {2014},

date = {2014-01-01},

urldate = {2014-01-01},

booktitle = {37th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO)},

pages = {1336--1341},

address = {Opatija, Croatia},

organization = {Mipro},

abstract = {Many techniques in the area of 3D face recognition rely on local descriptors to characterize the surface-shape information around points of interest (or keypoints) in the 3D images. Despite the fact that a lot of advancements have been made in the area of keypoint descriptors over the last years, the literature on 3D-face recognition for the most part still focuses on established descriptors, such as SIFT and SURF, and largely neglects more recent descriptors, such as the FREAK descriptor. In this paper we try to bridge this gap and assess the usefulness of the FREAK descriptor for the task for 3D face recognition. Of particular interest to us is a direct comparison of the FREAK and SIFT descriptors within a simple verification framework. To evaluate our framework with the two descriptors, we conduct 3D face recognition experiments on the challenging FRGCv2 and UMBDB databases and show that the FREAK descriptor ensures a very competitive verification performance when compared to the SIFT descriptor, but at a fraction of the computational cost. Our results indicate that the FREAK descriptor is a viable alternative to the SIFT descriptor for the problem of 3D face verification and due to its binary nature is particularly useful for real-time recognition systems and verification techniques for low-resource devices such as mobile phones, tablets and alike.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{vesnicer2014incorporating,

title = {Incorporating Duration Information into I-Vector-Based Speaker-Recognition Systems},

author = {Boštjan Vesnicer and Jerneja Žganec-Gros and Simon Dobrišek and Vitomir Štruc},

url = {https://lmi.fe.uni-lj.si/en/incorporatingdurationinformationintoi-vector-basedspeaker-recognitionsystems/},

year  = {2014},

date = {2014-01-01},

urldate = {2014-01-01},

booktitle = {Proceedings of Odyssey: The Speaker and Language Recognition Workshop},

pages = {241--248},

abstract = {Most of the existing literature on i-vector-based speaker recognition focuses on recognition problems, where i-vectors are extracted from speech recordings of sufficient length. The majority of modeling/recognition techniques therefore simply ignores the fact that the i-vectors are most likely estimated unreliably when short recordings are used for their computation. Only recently, were a number of solutions proposed in the literature to address the problem of duration variability, all treating the i-vector as a random variable whose posterior distribution can be parameterized by the posterior mean and the posterior covariance. In this setting the covariance matrix serves as a measure of uncertainty that is related to the length of the available recording. In contract to these solutions, we address the problem of duration variability through weighted statistics. We demonstrate in the paper how established feature transformation techniques regularly used in the area of speaker recognition, such as PCA or WCCN, can be modified to take duration into account. We evaluate our weighting scheme in the scope of the i-vector challenge organized as part of the Odyssey, Speaker and Language Recognition Workshop 2014 and achieve a minimal DCF of 0.280, which at the time of writing puts our approach in third place among all the participating institutions.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{marcetic2014experimental,

title = {An experimental tattoo de-identification system for privacy protection in still images},

author = {Darijan Marčetić and Slobodan Ribarić and Vitomir Štruc and Nikola Pavešić},

url = {https://lmi.fe.uni-lj.si/en/anexperimentaltattoode-identificationsystemforprivacyprotectioninstillimages/},

year  = {2014},

date = {2014-01-01},

urldate = {2014-01-01},

booktitle = {37th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO)},

pages = {1288--1293},

publisher = {IEEE},

organization = {Mipro},

abstract = {An experimental tattoo de-identification system for privacy protection in still images is described in the paper. The system consists of the following modules: skin detection, region of interest detection, feature extraction, tattoo database, matching, tattoo detection, skin swapping, and quality evaluation. Two methods for tattoo localization are presented. The first is a simple ad-hoc method based only on skin colour. The second is based on skin colour, texture and SIFT features. The appearance of each tattoo area is de-identified in such a way that its skin colour and skin texture are similar to the surrounding skin area. Experimental results for still images in which tattoo location, distance, size, illumination, and motion blur have large variability are presented. The system is subjectively evaluated based on the results of tattoo localization, the level of privacy protection and the naturalness of the de-identified still images. The level of privacy protection is estimated based on the quality of the removal of the tattoo appearance and the concealment of its location.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{beveridge2014ijcb,

title = {The ijcb 2014 pasc video face and person recognition competition},

author = {Ross Beveridge and Hao Zhang and Patrick Flynn and Yooyoung Lee and Venice Erin Liong and Jiwen Lu and Marcus Assis de Angeloni and Tiago Freitas de Pereira and Haoxiang Li and Gang Hua and Vitomir Štruc and Janez Križaj and Jonathon Phillips},

url = {https://lmi.fe.uni-lj.si/en/theijcb2014pascvideofaceandpersonrecognitioncompetition/},

year  = {2014},

date = {2014-01-01},

urldate = {2014-01-01},

booktitle = {Proceedings of the IEEE International Joint Conference on Biometrics (IJCB)},

pages = {1--8},

organization = {IEEE},

abstract = {The Point-and-Shoot Face Recognition Challenge (PaSC) is a performance evaluation challenge including 1401 videos of 265 people acquired with handheld cameras and depicting people engaged in activities with non-frontal head pose. This report summarizes the results from a competition using this challenge problem. In the Video-to-video Experiment a person in a query video is recognized by comparing the query video to a set of target videos. Both target and query videos are drawn from the same pool of 1401 videos. In the Still-to-video Experiment the person in a query video is to be recognized by comparing the query video to a larger target set consisting of still images. Algorithm performance is characterized by verification rate at a false accept rate of 0:01 and associated receiver operating characteristic (ROC) curves. Participants were provided eye coordinates for video frames. Results were submitted by 4 institutions: (i) Advanced Digital Science Center, Singapore; (ii) CPqD, Brasil; (iii) Stevens Institute of Technology, USA; and (iv) University of Ljubljana, Slovenia. Most competitors demonstrated video face recognition performance superior to the baseline provided with PaSC. The results represent the best performance to date on the handheld video portion of the PaSC.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{krizajrobust,

title = {Robust 3D face recognition using adapted statistical models},

author = {Janez Križaj and Simon Dobrišek and Vitomir Štruc and Nikola Pavešić},

url = {https://lmi.fe.uni-lj.si/en/robust3dfacerecognitionusingadaptedstatisticalmodels/},

year  = {2013},

date = {2013-09-20},

urldate = {2013-09-20},

booktitle = {Proceedings of the Electrotechnical and Computer Science Conference (ERK'13)},

abstract = {The paper presents a novel framework to 3D face recognition that exploits region covariance matrices (RCMs), Gaussian mixture models (GMMs) and support vector machine (SVM) classifiers. The proposed framework first combines several 3D face representations at the feature level using RCM descriptors and then derives low-dimensional feature vectors from the computed descriptors with the unscented transform. By doing so, it enables computations in Euclidean space, and makes Gaussian mixture modeling feasible. Finally, a support vector classifier is used for identity inference. As demonstrated by our experimental results on the FRGCv2 and UMB databases, the proposed framework is highly robust and exhibits desirable characteristics such as an inherent mechanism for data fusion (through the RCMs), the ability to examine local as well as global structures of the face with the same descriptor, the ability to integrate domain-specific prior knowledge into the modeling procedure and consequently to handle missing or unreliable data.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{ERK2013_Struc,

title = {Exploiting representation plurality for robust and efficient face recognition},

author = {Vitomir Štruc and Jeneja Žganec Gros and Simon Dobrišek and Nikola Pavešić},

url = {https://lmi.fe.uni-lj.si/en/exploitingrepresentationpluralityforrobustandefficientfacerecognition/},

year  = {2013},

date = {2013-09-01},

urldate = {2013-09-01},

booktitle = {Proceedings of the 22nd Intenational Electrotechnical and Computer Science Conference (ERK'13)},

volume = {vol. B},

pages = {121--124},

address = {Portorož, Slovenia},

abstract = {The paper introduces a novel approach to face recognition that exploits plurality of representation to achieve robust face recognition. The proposed approach was submitted as a representative of the University of Ljubljana and Alpineon d.o.o. to the 2013 face recognition competition that was held in conjunction with the IAPR International Conference on Biometrics and achieved the best overall recognition results among all competition participants. Here, we describe the basic characteristics of the submitted approach, elaborate on the results of the competition and, most importantly, present some general findings made during our development work that are of relevance to the broader (face recognition) research community.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{FG2013,

title = {Combining 3D face representations using region covariance descriptors and statistical models},

author = {Janez Križaj and Vitomir Štruc and Simon Dobrišek},

url = {https://lmi.fe.uni-lj.si/en/combining3dfacerepresentationsusingregioncovariancedescriptorsandstatisticalmodels/},

year  = {2013},

date = {2013-05-01},

urldate = {2013-05-01},

booktitle = {Proceedings of the IEEE International Conference on Automatic Face and Gesture Recognition and Workshops (IEEE FG), Workshop on 3D Face Biometrics},

publisher = {IEEE},

address = {Shanghai, China},

abstract = {The paper introduces a novel framework for 3D face recognition that capitalizes on region covariance descriptors and Gaussian mixture models. The framework presents an elegant and coherent way of combining multiple facial representations, while simultaneously examining all computed representations at various levels of locality. The framework first computes a number of region covariance matrices/descriptors from different sized regions of several image representations and then adopts the unscented transform to derive low-dimensional feature vectors from the computed descriptors. By doing so, it enables computations in the Euclidean space, and makes Gaussian mixture modeling feasible. In the last step a support vector machine classification scheme is used to make a decision regarding the identity of the modeled input 3D face image. The proposed framework exhibits several desirable characteristics, such as an inherent mechanism for data fusion/integration (through the region covariance matrices), the ability to examine the facial images at different levels of locality, and the ability to integrate domain-specific prior knowledge into the modeling procedure. We assess the feasibility of the proposed framework on the Face Recognition Grand Challenge version 2 (FRGCv2) database with highly encouraging results.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{vstruc2013patch,

title = {Patch-wise low-dimensional probabilistic linear discriminant analysis for Face Recognition},

author = {Vitomir Štruc and Nikola Pavešić and Jerneja Žganec-Gros and Boštjan Vesnicer},

url = {https://lmi.fe.uni-lj.si/en/patch-wiselow-dimensionalprobabilisticlineardiscriminantanalysisforfacerecognition/},

doi = {10.1109/ICASSP.2013.6638075},

year  = {2013},

date = {2013-01-01},

urldate = {2013-01-01},

booktitle = {2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},

pages = {2352--2356},

organization = {IEEE},

abstract = {The paper introduces a novel approach to face recognition based on the recently proposed low-dimensional probabilistic linear discriminant analysis (LD-PLDA). The proposed approach is specifically designed for complex recognition tasks, where highly nonlinear face variations are typically encountered. Such data variations are commonly induced by changes in the external illumination conditions, viewpoint changes or expression variations and represent quite a challenge even for state-of-the-art techniques, such as LD-PLDA. To overcome this problem, we propose here a patch-wise form of the LDPLDA technique (i.e., PLD-PLDA), which relies on local image patches rather than the entire image to make inferences about the identity of the input images. The basic idea here is to decompose the complex face recognition problem into simpler problems, for which the linear nature of the LD-PLDA technique may be better suited. By doing so, several similarity scores are derived from one facial image, which are combined at the final stage using a simple sum-rule fusion scheme to arrive at a single score that can be employed for identity inference. We evaluate the proposed technique on experiment 4 of the Face Recognition Grand Challenge (FRGCv2) database with highly promising results.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{gunther20132013,

title = {The 2013 face recognition evaluation in mobile environment},

author = {Manuel Günther and Artur Costa-Pazo and Changxing Ding and Elhocine Boutellaa and Giovani Chiachia and Honglei Zhang and Marcus Assis de Angeloni and Vitomir Štruc and Elie Khoury and Esteban Vazquez-Fernandez and others},

url = {https://lmi.fe.uni-lj.si/en/the2013facerecognitionevaluationinmobileenvironment/},

year  = {2013},

date = {2013-01-01},

urldate = {2013-01-01},

booktitle = {Proceedings of the IAPR International Conference on Biometrics (ICB)},

pages = {1--7},

organization = {IAPR},

abstract = {Automatic face recognition in unconstrained environments is a challenging task. To test current trends in face recognition algorithms, we organized an evaluation on face recognition in mobile environment. This paper presents the results of 8 different participants using two verification metrics. Most submitted algorithms rely on one or more of three types of features: local binary patterns, Gabor wavelet responses including Gabor phases, and color information. The best results are obtained from UNILJ-ALP, which fused several image representations and feature types, and UCHU, which learns optimal features with a convolutional neural network. Additionally, we assess the usability of the algorithms in mobile devices with limited resources.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{BioID_Struc_2011,

title = {Principal directions of synthetic exact filters for robust real-time eye localization},

author = {Vitomir Štruc and Jerneja Žganec-Gros and Nikola Pavešić},

url = {https://lmi.fe.uni-lj.si/en/principaldirectionsofsyntheticexactfiltersforrobustreal-timeeyelocalization/},

doi = {10.1007/978-3-642-19530-3_17},

year  = {2011},

date = {2011-01-01},

urldate = {2011-01-01},

booktitle = {Proceedings of the COST workshop on Biometrics and Identity Management (BioID)},

volume = {6583/2011},

pages = {180/192},

publisher = {Springer-Verlag},

address = {Berlin, Heidelberg},

series = {Lecture Notes on Computer Science},

abstract = {The alignment of the facial region with a predefined canonical form is one of the most crucial steps in a face recognition system. Most of the existing alignment techniques rely on the position of the eyes and, hence, require an efficient and reliable eye localization procedure. In this paper we propose a novel technique for this purpose, which exploits a new class of correlation filters called Principal directions of Synthetic Exact Filters (PSEFs). The proposed filters represent a generalization of the recently proposed Average of Synthetic Exact Filters (ASEFs) and exhibit desirable properties, such as relatively short training times, computational simplicity, high localization rates and real time capabilities. We present the theory of PSEF filter construction, elaborate on their characteristics and finally develop an efficient procedure for eye localization using several PSEF filters. We demonstrate the effectiveness of the proposed class of correlation filters for the task of eye localization on facial images from the FERET database and show that for the tested task they outperform the established Haar cascade object detector as well as the ASEF correlation filters.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}