The challenge of face recognition from digital point-and-shoot cameras

Beveridge, J. Ross; Phillips, P. Jonathon; Bolme, David S.; Draper, Bruce A.; Givens, Geof H.; Lui, Yui Man; Teli, Mohammad Nayeem; Zhang, Hao; Scruggs, William; Bowyer, Kevin W.; Flynn, Patrick J.; Cheng, Su

doi:10.1109/btas.2013.6712704

Cited by 161 publications

(132 citation statements)

References 22 publications

Supporting

Mentioning

131

Contrasting

Order By: Relevance

“…These video surveillance datasets can be employed to emulate real-world watch-list screening applications. The main characteristics of these two datasets with respect to others [11], [30] are that they contain a high-quality still face images captured under controlled condition (with the same still camera), and low-quality surveillance videos for each subject captured under uncontrolled conditions (with surveillance cameras). Chokepoint dataset can be used as a benchmark for large-scale FR, especially in watch-list applications.…”

Section: Video Databasementioning

confidence: 99%

Robust watch-list screening using dynamic ensembles of SVMs based on multiple face representations

Bashbaghi

Granger

Sabourin

et al. 2017

Machine Vision and Applications

View full text Add to dashboard Cite

Abstract. Still-to-video face recognition (FR) is an important function in video surveillance, where faces captured over a network of video cameras are matched against reference stills of target individuals. Screening faces against a watch-list is a challenging video surveillance application because the appearance of faces vary due to changing capture conditions and operational domains. The facial models used for matching may not be representative of faces captured with video cameras because they are typically designed a priori with only one reference still. In this paper, a multi-classifier framework is proposed for robust still-to-video FR based on multiple and diverse face representations of a single reference face still. During enrollment of a target individual, the single reference face still is modeled using an ensemble of SVM classifiers based on different patches and face descriptors. Multiple feature extraction techniques are applied to patches isolated in the reference still to generate a diverse SVM pool that provides robustness to common nuisance factors (e.g., variations in illumination and pose). The estimation of discriminant feature subsets, classifier parameters, decision thresholds, and ensemble fusion functions is achieved using the high-quality reference still and a large number of faces captured in lower quality video of non-target individuals in the scene. During operations, the most competent subset of SVMs are dynamically selected according to capture conditions. Finally, a head-face tracker gradually regroups faces captured from different people appearing in a scene, while each individual-specific ensemble performs face matching. The accumulation of matching scores per face track leads to a robust spatio-temporal FR when accumulated ensemble scores surpass a detection threshold. Experimental results obtained with the Chokepoint and COX-S2V datasets show a significant improvement in performance w.r.t. reference systems, especially when individual-specific ensembles (1) are designed using exemplar-SVMs rather than one-class SVMs, and (2) exploit score-level fusion of local SVMs (trained using features extracted from each patch), rather than using either decision-level or feature-level fusion with a global SVM (trained by concatenating features extracted from patches).

show abstract

Section: Video Databasementioning

confidence: 99%

Robust watch-list screening using dynamic ensembles of SVMs based on multiple face representations

Bashbaghi

Granger

Sabourin

et al. 2017

Machine Vision and Applications

View full text Add to dashboard Cite

show abstract

“…The videos in the PaSC were acquired in seven weeks spread out over the Spring 2011 academic semester at the 2 The identification of any commercial product or trade name does not imply endorsement or recommendation by NIST. University of Notre Dame.…”

Section: A Video Datamentioning

confidence: 99%

“…Section III provides additional background on the PaSC [2], the two experiments included in this evaluation, and the evaluation protocol. Next, in Section IV, the approaches taken by each of the five participants are summarized.…”

Section: Introductionmentioning

confidence: 99%

“…By design, many of the complications that arise in video face recognition are amply represented in the Point-andShoot Challenge Face Recognition Challenge (PaSC) [2]; the FG 2015 Video Person Recognition Evaluation consists of two experiments from the PaSC. The videos in the PaSC data set show people in motion carrying out actions; the goal is to recognize the people performing the actions, not to recognize the actions.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Report on the FG 2015 Video Person Recognition Evaluation

Beveridge

Zhang

Draper

et al. 2015

2015 11th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition (FG)

Self Cite

View full text Add to dashboard Cite

show abstract

“…For determining the position to split the videos, we use the mean position of the camera positions, as described in section 4.5.1, in the evaluation section. Recognition Challenge (PaSC) [2] was created for two reasons:…”

Section: Splitting the Video Into Segmentsmentioning

confidence: 99%

Video alignment to a common reference

Dutta

Draper

Beveridge

2014

IEEE Winter Conference on Applications of Computer Vision

Self Cite

View full text Add to dashboard Cite

Video Alignment to a Common ReferenceHandheld videos often include unintentional motion (jitter) and intentional motion (pan and/or zoom). Human viewers prefer to see jitter removed, creating a smoothly moving camera. For video analysis, in contrast, aligning to a fixed stable background is sometimes preferable. This paper presents an algorithm that removes both forms of motion using a novel and efficient way of tracking background points while ignoring moving foreground points.The approach is related to image mosaicing, but the result is a video rather than an enlarged still image. It is also related to multiple object tracking approaches, but simpler since moving objects need not be explicitly tracked. The algorithm presented takes as input a video and returns one or several stabilized videos. Videos are broken into parts when the algorithm detects background change and it becomes necessary to fix upon a new background. We present two techniques in this thesis. One technique stabilizes the video with respect to the first available frame. Another technique stabilizes the videos with respect to a best frame. Our approach assumes the person holding the camera is standing in one place and that objects in motion do not dominate the image. Our algorithm performs better than previously published approaches when compared on 1,401 handheld videos from the recently released Point-and-Shoot Face Recognition Challenge (PASC).ii

show abstract

The challenge of face recognition from digital point-and-shoot cameras

Cited by 161 publications

References 22 publications

Robust watch-list screening using dynamic ensembles of SVMs based on multiple face representations

Robust watch-list screening using dynamic ensembles of SVMs based on multiple face representations

Report on the FG 2015 Video Person Recognition Evaluation

Video alignment to a common reference

Contact Info

Product

Resources

About