MagnifierNet: Towards Semantic Adversary and Fusion for Person Re-identification
arXiv:2002.10979
Abstract
Although person re-identification (ReID) has achieved significant improvement recently by enforcing part alignment, it is still a challenging task when it comes to distinguishing visually similar identities or identifying the occluded person. In these scenarios, magnifying details in each part features and selectively fusing them together may provide a feasible solution. In this work, we propose MagnifierNet, a triple-branch network which accurately mines details from whole to parts. Firstly, the holistic salient features are encoded by a global branch. Secondly, to enhance detailed representation for each semantic region, the "Semantic Adversarial Branch" is designed to learn from dynamically generated semantic-occluded samples during training. Meanwhile, we introduce "Semantic Fusion Branch" to filter out irrelevant noises by selectively fusing semantic region information sequentially. To further improve feature diversity, we introduce a novel loss function "Semantic Diversity Loss" to remove redundant overlaps across learned semantic representations. State-of-the-art performance has been achieved on three benchmarks by large margins. Specifically, the mAP score is improved by 6% and 5% on the most challenging CUHK03-L and CUHK03-D benchmarks.
References in corpus (8)
- In Defense of the Triplet Loss for Person Re-Identification
- On the Properties of Neural Machine Translation: Encoder-Decoder Approaches
- A Discriminatively Learned CNN Embedding for Person Re-identification
- AlignedReID: Surpassing Human-Level Performance in Person Re-Identification
- Omni-Scale Feature Learning for Person Re-Identification
- Foreground-aware Pyramid Reconstruction for Alignment-free Occluded Person Re-identification
- EANet: Enhancing Alignment for Cross-Domain Person Re-identification
- Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training