3,646 Hits in 4.1 sec

Self-Updating with Facial Trajectories for Video-to-Video Face Recognition

Miguel De-la-Torre, Eric Granger, Paulo Radtke, Robert Sabourin, Dmitry O. Gorodnichy
2014 2014 22nd International Conference on Pattern Recognition  
In this paper, an adaptive MCS is proposed that allows for self-updating facial models given face trajectories captured during operations.  ...  For applications of face recognition (FR) in video surveillance, it is often costly or unfeasible to collect several high quality reference samples a priori to design representative facial models.  ...  ACKNOWLEDGMENT This work was partially supported by the Natural Sciences and Engineering Research Council of Canada, and the Mexican National Council for Science and Technology.  ... 
doi:10.1109/icpr.2014.294 dblp:conf/icpr/De-la-TorreGRSG14 fatcat:q34wbjbk5vawjduj7ntjb4vq2m

Adaptive appearance model tracking for still-to-video face recognition

M. Ali Akber Dewan, E. Granger, G.-L. Marcialis, R. Sabourin, F. Roli
2016 Pattern Recognition  
Systems for still-to-video face recognition (FR) seek to detect the presence of target individuals based on reference facial still images or mug-shots.  ...  Compared to state-ofthe-art adaptive biometric systems, the proposed method selects facial captures to update an individual's face model more reliably because it relies on information from tracking.  ...  Supplementary data Supplementary data associated with this paper can be found in the online version at  ... 
doi:10.1016/j.patcog.2015.08.002 fatcat:wexkxdb25zgflpb5uzulq53x34

An adaptive ensemble-based system for face recognition in person re-identification

Miguel De-la-Torre, Eric Granger, Robert Sabourin, Dmitry O. Gorodnichy
2015 Machine Vision and Applications  
Tracking can, however, be used to regroup the system responses linked to a facial trajectory (facial captures from a person) for robust spatio-temporal recognition, and to update facial models over time  ...  During operations, B Miguel De-la-Torre facial trajectories are captured, and each individual-specific ensemble of the system seeks to detect target individuals, and possibly self-update their facial models  ...  This work was also supported by the Program for the Improvement of the Professoriate of the Secretariat of Public Education, Mexico, the Mexican National Council for Science and Technology, and the University  ... 
doi:10.1007/s00138-015-0697-7 fatcat:dyy7iwpnlvfr5f4i3jmukb4nbm

Video Face Recognition From A Single Still Image Using an Adaptive Appearance Model Tracker

M. Ali Akber Dewan, E. Granger, R. Sabourin, G.L. Marcialis, F. Roli
2015 2015 IEEE Symposium Series on Computational Intelligence  
Systems for still-to-video face recognition (FR) are typically used to detect target individuals in watch-list screening applications.  ...  Track-facemodels allow selecting facial captures for modeling more reliably than self-updating FR systems, and can incorporate a greater diversity of intra-class variation from the operational environment  ...  Self-updating is limited on a single trait (e.g., face) to update the GFM of a target individual.  ... 
doi:10.1109/ssci.2015.38 dblp:conf/ssci/DewanGSMR15 fatcat:j2gl4mrvkvairgie7hnu4tsgxq

Partially-supervised learning from facial trajectories for face recognition in video surveillance

Miguel De-la-Torre, Eric Granger, Paulo V.W. Radtke, Robert Sabourin, Dmitry O. Gorodnichy
2015 Information Fusion  
During operations, information from a face tracker and individual-specific ensembles is integrated for robust spatio-temporal recognition and for self-update of facial models.  ...  Subject-level analysis reveals the existence of individuals for which self-updating ensembles with unlabeled facial trajectories provides a considerable benefit.  ...  This work was also supported by the Program for the Improvement of the Professoriate of the Secretariat of Public Education, Mexico, and the Mexican National Council for Science and Technology.  ... 
doi:10.1016/j.inffus.2014.05.006 fatcat:wudpapepanbs7k4yberva6f6pa

Individual-specific management of reference data in adaptive ensembles for face re-identification

Miguel De-la-Torre, Dmitry O. Gorodnichy, Eric Granger, Robert Sabourin
2015 IET Computer Vision  
In addition, facial models for recognition are commonly designed using a limited number of representative reference samples from faces captured under specific conditions, regrouped into facial trajectories  ...  Only the samples with the highest KL divergence are preserved in memory for future adaptations. This strategy is compared with reference classifiers using videos from the face in action data.  ...  This work was also supported by the Program for the Improvement of the Professoriate of the Secretariat of Public Education, Mexico, and the Mexican National Council for Science and Technology.  ... 
doi:10.1049/iet-cvi.2014.0375 fatcat:dgygeg6esjahzn22jizt6qxani

Correspondence driven adaptation for human profile recognition

Ming Yang, Shenghuo Zhu, Fengjun Lv, Kai Yu
2011 CVPR 2011  
We enforce the system to output consistent and stable results on face images from the same trajectories in videos by using incremental stochastic training.  ...  Visual recognition systems for videos using statistical learning models often show degraded performance when being deployed to a real-world environment, primarily due to the fact that training data can  ...  For race recognition, we test on a dataset with 908 persons and 19858 faces collected from video clips recorded in a shopping mall, thus, referred as the Mall dataset.  ... 
doi:10.1109/cvpr.2011.5995481 dblp:conf/cvpr/YangZLY11 fatcat:ujc4tourfza33o4qhc3edsm5dy

Surveillance using Face Recognition in Smart Cities

In this project safe city demonstrates how the security in India can be increased with the help of video surveillance using facial recognition.  ...  But now it's time to build safe cities for India.  ...  For robust spatio-temporal recognition and for facial model self-updating, information from a face tracker and individual-specific ensembles is integrated during operations.  ... 
doi:10.35940/ijitee.f3582.049620 fatcat:rgzpkd42krfddjapt66djuetjq

Adaptive skew-sensitive fusion of ensembles and their application to face re-identification

Miguel De-la-Torre, Eric Granger, Robert Sabourin
2015 2015 International Joint Conference on Neural Networks (IJCNN)  
Adaptive classifier ensembles have been shown to improve the accuracy and robustness of systems for face recognition (FR) in video surveillance.  ...  During enrollment, faces in a reference trajectory are combined with those of selected non-target trajectories to generate a pool of 2-class classifiers using data with various levels of imbalance and  ...  Given a video surveillance system, an analyst may capture a reference facial trajectory corresponding to a target individual appearing in video feeds, and then design or update a facial model (e.g., a  ... 
doi:10.1109/ijcnn.2015.7280433 dblp:conf/ijcnn/De-la-TorreGS15 fatcat:zygwirlnrjcydl2rks6ymve25m

Development of a new 3D tracking system for multiple marmosets under free-moving conditions [article]

Terumi Yurimoto, Wakako Kumita, Kenya Sato, Rika Kikuchi, Yusuke Shibuki, Rino Hashimoto, Michiko Kamioka, Yumi Hayasegawa, Eiko Yamazaki, Yoko Kurotaki, Norio Goda, Junichi Kitakami (+3 others)
2022 bioRxiv   pre-print
Their trajectories were tracked independently by combining video, 3D coordinates detected by light detection and ranging (lidar) systems, and facial identification.  ...  Each marmoset was identified using deep learning facial recognition (accuracy ≥ 97%).  ...  To match the tracking trajectory with the individual animal information, facial recognition was performed to identify individuals.  ... 
doi:10.1101/2022.03.29.486138 fatcat:4lh46cpztzdl7it7fiavjplnta

On the Effects of Illumination Normalization with LBP-Based Watchlist Screening [chapter]

Ibtihel Amara, Eric Granger, Abdenour Hadid
2015 Lecture Notes in Computer Science  
Still-to-video face recognition (FR) is an important function in several video surveillance applications like watchlist screening, where faces captured over a network of video cameras are matched against  ...  This study focuses on representative retinex, self-quotient, diffusion, filtering, means de-noising, retina, wavelet and frequency-based techniques that are suitable for fast and accurate face screening  ...  Videos are challenging for still-to video FR since faces are captured under semi-controlled conditions, with changes in illumination, pose, scale, blur and occlusion.  ... 
doi:10.1007/978-3-319-16181-5_13 fatcat:lnapgsu5ezfxdnf3dzcghgn37m

Five-Month-old Infants' Discrimination of Visual-Tactile Synchronous Facial Stimulation

M. L. Filippetti, T. Farroni, M. H. Johnson
2016 Infant and Child Development  
However, this use of a peer face also prevented us from systematically exploring the role of familiarity with own personal facial features in the discrimination between self and other (as in the enfacement  ...  effective not only with limbs, but also with the most representative component of self-identity, that is the face (Tajadura-Jimenez et al., 2012b).  ... 
doi:10.1002/icd.1977 fatcat:tywjtl3i2faelpxafdopvpgpui

To Frontalize or Not To Frontalize: Do We Really Need Elaborate Pre-processing To Improve Face Recognition? [article]

Sandipan Banerjee, Joel Brogan, Janez Krizaj, Aparna Bharati, Brandon RichardWebster, Vitomir Struc, Patrick Flynn, Walter Scheirer
2018 arXiv   pre-print
To address this question, we evaluate a number of popular facial landmarking and pose correction algorithms to understand their effect on facial recognition performance.  ...  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  ...  Methodology To analyze the effect of facial frontalization on recognition performance, we trained the VGG-FACE network separately for each subset of training data pre-processed with a given method.  ... 
arXiv:1610.04823v4 fatcat:dqwknkd4cvd6xl2wrosu4h6f2e

Dynamic multi-objective evolution of classifier ensembles for video face recognition

Jean-François Connolly, Eric Granger, Robert Sabourin
2013 Applied Soft Computing  
Due to a limited control over changing operational conditions and personal physiology, systems used for video-based face recognition are confronted with complex and changing pattern recognition environments  ...  For real time recognition of individuals appearing in video sequences, facial regions are captured with one or more cameras, and an AMCS must perform fast and efficient matching against the facial model  ...  This paper seeks to address challenges related to the design of robust adaptive multi-classifier systems (AMCSs) for video face recognition, where facial models may be created and updated over time, as  ... 
doi:10.1016/j.asoc.2012.08.039 fatcat:nc5t4ap75req7fz7aka6mjjwri

Adaptive ensembles for face recognition in changing video surveillance environments

C. Pagano, E. Granger, R. Sabourin, G.L. Marcialis, F. Roli
2014 Information Sciences  
Face recognition (FR) systems are exposed to videos captured under various operating conditions, and, since data distributions change over time, face captures diverge w.r.t. stored facial models.  ...  By adapting facial models according to changes detected in new reference videos, this AMCS allows to sustain a high level of accuracy comparable to the same system that is always updated using a learn-and-combine  ...  A generic system for video face recognition Fig. 1 presents a generic system for video-to-video FR.  ... 
doi:10.1016/j.ins.2014.07.005 fatcat:x74ha7svqvhwfjqny76jwyfswm
« Previous Showing results 1 — 15 out of 3,646 results