Feature-Level Frankenstein: Eliminating Variations for Discriminative Recognition

Xiaofeng Liu, Site Li, Lingsheng Kong, Wanqing Xie, Ping Jia, Jane You, B.V.K. Kumar; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 637-646

Abstract


Recent successes of deep learning-based recognition rely on maintaining the content related to the main-task label. However, how to explicitly dispel the noisy signals for better generalization remains an open issue. We systematically summarize the detrimental factors as task-relevant/irrelevant semantic variations and unspecified latent variation. In this paper, we cast these problems as an adversarial minimax game in the latent space. Specifically, we propose equipping an end-to-end conditional adversarial network with the ability to decompose an input sample into three complementary parts. The discriminative representation inherits the desired invariance property guided by prior knowledge of the task, which is marginally independent to the task-relevant/irrelevant semantic and latent variations. Our proposed framework achieves top performance on a serial of tasks, including digits recognition, lighting, makeup, disguise-tolerant face recognition, and facial attributes recognition.

Related Material


[pdf]
[bibtex]
@InProceedings{Liu_2019_CVPR,
author = {Liu, Xiaofeng and Li, Site and Kong, Lingsheng and Xie, Wanqing and Jia, Ping and You, Jane and Kumar, B.V.K.},
title = {Feature-Level Frankenstein: Eliminating Variations for Discriminative Recognition},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}