Unconstrained Still/Video-Based Face Verification with Deep Convolutional Neural Networks

Abstract

Over the last five years, methods based on Deep Convolutional Neural Networks (DCNNs) have shown impressive performance improvements for object detection and recognition problems. This has been made possible due to the availability of large annotated datasets, a better understanding of the non-linear mapping between input images and class labels as well as the affordability of GPUs. In this paper, we present the design details of a deep learning system for unconstrained face recognition, including modules for face detection, association, alignment and face verification. The quantitative performance evaluation is conducted using the IARPA Janus Benchmark A (IJB-A), the JANUS Challenge Set 2 (JANUS CS2), and the LFW dataset. The IJB-A dataset includes real-world unconstrained faces of 500 subjects with significant pose and illumination variations which are much harder than the Labeled Faces in the Wild (LFW) and Youtube Face (YTF) datasets. JANUS CS2 is the extended version of IJB-A which contains not only all the images/frames of IJB-A but also includes the original videos. Some open issues regarding DCNNs for face verification problems are then discussed.

Video Face Recognition: Component-wise Feature Aggregation Network (C-FAN)

“…IJB-A contains 500 subjects with a total of 25, 813 images, and has been widely used by a number of both still image and video-based face recognition algorithms. 4 The "booking" reference template comprises the full set of images captured of a single subject at enrollment time. 5 UAV is a small fixed-wing unmanned aerial vehicle that was flown to collect images and videos.…”

OGCTL: Occlusion-guided compact template learning for ensemble deep network-based pose-invariant face recognition

“…(ii) the templates generated from an Templates t generated from either architecture are sensitive to non-discriminative facial representations x i+n generated from occluded facial patches. (L) Network architecture employed in [32,30,31,2]. (R) Network architecture employed in [24].…”

Using CNN's Gait Recognition to Strengthen Laboratory Safety Supervision

“…For example, Zhang et al introduced an image denoising task using DnCNN model [20]. Chen et al proposed an unconstrained algorithm using Deep CNN for feature-based human connection verification, obtaining good experimental results [21]. Their paper also refers to the model by Alotaibi et al, using a self-made Deep CNN architecture to complete the identification of the CASIA gait database with different viewing angles.…”