TY - GEN
T1 - Confidence-Driven Network for Point-to-Set Matching
AU - Leng, Mengjun
AU - Kakadiaris, Ioannis A.
N1 - Funding Information:
In this paper, we proposed the Confidence Driven Network (CDN): a framework that jointly learns a feature vector and a confidence score for each image in a multi-probe face identification setup. To learn the confidence score, a single sample-test mechanism was introduced to quantify the discriminative level of the template. CDN improves the rank-1 identification rate for multi-probe face identification in the selected datasets. We observed that CDN exhibits superior performance when image sets contain large pose variations, whereas for large image sets, the improvements are not significant. We identified several visual properties of the original image (e.g., pose, illumination, skin color) that affect the confidence score and provided quantitative and qualitative results to support our claims. Which exactly are these properties and to what extent they contribute to the identification accuracy is a topic for future research. Acknowledgement: This material is based upon work supported by the U.S. Department of Homeland Security under Grant Award Number 2015-ST-061-BSH001. This grant is awarded to the Borders, Trade, and Immigration (BTI) Institute: A DHS Center of Excellence led by the University of Houston, and includes support for the project Image and Video Person Identification in an Operational Environment: Phase I awarded to the University of Houston. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of the U.S. Department of Homeland Security
Publisher Copyright:
© 2018 IEEE.
PY - 2018/11/26
Y1 - 2018/11/26
N2 - The goal of point-to-set matching is to match a single image with a set of images from a subject. Within an image set, different images contain various levels of discriminative information and thus should contribute differently to the results. However, the discriminative level is not accessible directly. To this end, we propose a confidence driven network to perform point-to-set matching. The proposed system comprises a feature extraction network (FEN) and a performance prediction network (PPN). Given an input image, the FEN generates a template, while the PPN generates a confidence score which measures the discriminative level of the template. At matching time, the template is used to compute a point-to-point similarity. The similarity scores from different samples in the set are integrated at a score level, weighted by the predicted confidence scores. Extensive multi-probe face recognition experiments on the IJB-A and UHDB-31 datasets demonstrate performance improvements over state of the art algorithms.
AB - The goal of point-to-set matching is to match a single image with a set of images from a subject. Within an image set, different images contain various levels of discriminative information and thus should contribute differently to the results. However, the discriminative level is not accessible directly. To this end, we propose a confidence driven network to perform point-to-set matching. The proposed system comprises a feature extraction network (FEN) and a performance prediction network (PPN). Given an input image, the FEN generates a template, while the PPN generates a confidence score which measures the discriminative level of the template. At matching time, the template is used to compute a point-to-point similarity. The similarity scores from different samples in the set are integrated at a score level, weighted by the predicted confidence scores. Extensive multi-probe face recognition experiments on the IJB-A and UHDB-31 datasets demonstrate performance improvements over state of the art algorithms.
UR - http://www.scopus.com/inward/record.url?scp=85059785601&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85059785601&partnerID=8YFLogxK
U2 - 10.1109/ICPR.2018.8545036
DO - 10.1109/ICPR.2018.8545036
M3 - Conference contribution
AN - SCOPUS:85059785601
T3 - Proceedings - International Conference on Pattern Recognition
SP - 3414
EP - 3420
BT - 2018 24th International Conference on Pattern Recognition, ICPR 2018
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 24th International Conference on Pattern Recognition, ICPR 2018
Y2 - 20 August 2018 through 24 August 2018
ER -