Person Re-identification with Adversarial Triplet Embedding
This paper offers an incremental improvement to person re-identification, which is a problem for public security and video surveillance applications.
This paper addresses the problem of person re-identification, which is crucial for video surveillance. The authors propose Adversarial Triplet Embedding (ATE) to overcome the limitations of traditional triplet loss, specifically poor local optima and reliance on hard example mining, by generating adversarial triplets and discriminative feature embeddings simultaneously. The method demonstrates effectiveness on several benchmark datasets.
Person re-identification is an important task and has widespread applications in video surveillance for public security. In the past few years, deep learning network with triplet loss has become popular for this problem. However, the triplet loss usually suffers from poor local optimal and relies heavily on the strategy of hard example mining. In this paper, we propose to address this problem with a new deep metric learning method called Adversarial Triplet Embedding (ATE), in which we simultaneously generate adversarial triplets and discriminative feature embedding in an unified framework. In particular, adversarial triplets are generated by introducing adversarial perturbations into the training process. This adversarial game is converted into a minimax problem so as to have an optimal solution from the theoretical view. Extensive experiments on several benchmark datasets demonstrate the effectiveness of the approach against the state-of-the-art literature.