Learning Deep Features via Congenerous Cosine Loss for Person Recognition

TL;DR

This paper introduces a novel congenerous cosine loss function for person recognition that improves feature robustness and discrimination by directly optimizing cosine distances, leading to higher accuracy.

Contribution

The paper proposes a new congenerous cosine loss that simplifies training and enhances inter-class separation without additional test-time training.

Findings

Achieves higher classification accuracy than previous methods.

Effectively reduces intra-class variance.

Simplifies training process with cosine-based loss.

Abstract

Person recognition aims at recognizing the same identity across time and space with complicated scenes and similar appearance. In this paper, we propose a novel method to address this task by training a network to obtain robust and representative features. The intuition is that we directly compare and optimize the cosine distance between two features - enlarging inter-class distinction as well as alleviating inner-class variance. We propose a congenerous cosine loss by minimizing the cosine distance between samples and their cluster centroid in a cooperative way. Such a design reduces the complexity and could be implemented via softmax with normalized inputs. Our method also differs from previous work in person recognition that we do not conduct a second training on the test subset. The identity of a person is determined by measuring the similarity from several body regions in the reference set. Experimental results show that the proposed approach achieves better classification accuracy against previous state-of-the-arts.