Abstract | ||
---|---|---|
We propose a method to push the frontiers of unconstrained face recognition in the wild, focusing on the problem of extreme pose variations. As opposed to current techniques which either expect a single model to learn pose invariance through massive amounts of training data, or which normalize images to a single frontal pose, our method explicitly tackles pose variation by using multiple pose specific models and rendered face images. We leverage deep Convolutional Neural Networks (CNNs) to learn discriminative representations we call Pose-Aware Models (PAMs) using 500K images from the CASIA WebFace dataset. We present a comparative evaluation on the new IARPA Janus Benchmark A (IJB-A) and PIPA datasets. On these datasets PAMs achieve remarkably better performance than commercial products and surprisingly also outperform methods that are specifically fine-tuned on the target dataset. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1109/CVPR.2016.523 | 2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) |
Field | DocType | Volume |
Training set,Computer vision,Facial recognition system,Normalization (statistics),Three-dimensional face recognition,Pattern recognition,Computer science,Convolutional neural network,Artificial intelligence,Discriminative model,Machine learning | Conference | 2016 |
Issue | ISSN | Citations |
1 | 1063-6919 | 17 |
PageRank | References | Authors |
0.59 | 13 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Iacopo Masi | 1 | 348 | 16.19 |
stephen rawls | 2 | 59 | 4.08 |
Gérard G. Medioni | 3 | 2399 | 255.72 |
Premkumar Natarajan | 4 | 874 | 79.46 |