Unconstrained Face Verification using Deep CNN Features
About
In this paper, we present an algorithm for unconstrained face verification based on deep convolutional features and evaluate it on the newly released IARPA Janus Benchmark A (IJB-A) dataset. The IJB-A dataset includes real-world unconstrained faces from 500 subjects with full pose and illumination variations which are much harder than the traditional Labeled Face in the Wild (LFW) and Youtube Face (YTF) datasets. The deep convolutional neural network (DCNN) is trained using the CASIA-WebFace dataset. Extensive experiments on the IJB-A dataset are provided.
Jun-Cheng Chen, Vishal M. Patel, Rama Chellappa• 2015
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Face Verification | LFW | -- | 339 | |
| Face Verification | LFW (Labeled Faces in the Wild) unrestricted-labeled-outside-data protocol 14 | Accuracy97.45 | 47 | |
| Face Search | IJB-A | Rank@190.3 | 44 | |
| Face Verification | IJB-A | TAR @ FAR=1%0.838 | 38 | |
| Face Verification | IJB-A (test) | TAR @ FAR=0.010.967 | 37 | |
| Face Identification | IJB-A (test) | Rank-190.3 | 30 | |
| Face Recognition | CS2 | Rank-1 Accuracy89.1 | 21 | |
| Face Verification | IJB-A (10 folds average) | TAR @ FAR=0.0183.8 | 18 | |
| Face Recognition | IJB-A (test) | TAR @ FAR=0.0183.8 | 16 | |
| Closed-set Face Search | IJB-A (10 folds average) | Rank-10.903 | 9 |
Showing 10 of 11 rows