Identity-aware Feature Decoupling Learning for Clothing-change Person Re-identification
This addresses the challenge of identifying individuals across different clothing in surveillance applications, representing an incremental improvement over existing methods.
The paper tackles the problem of clothing-change person re-identification by proposing an Identity-aware Feature Decoupling learning framework to better extract identity-related features, achieving superior performance on multiple datasets compared to baseline models.
Clothing-change person re-identification (CC Re-ID) has attracted increasing attention in recent years due to its application prospect. Most existing works struggle to adequately extract the ID-related information from the original RGB images. In this paper, we propose an Identity-aware Feature Decoupling (IFD) learning framework to mine identity-related features. Particularly, IFD exploits a dual stream architecture that consists of a main stream and an attention stream. The attention stream takes the clothing-masked images as inputs and derives the identity attention weights for effectively transferring the spatial knowledge to the main stream and highlighting the regions with abundant identity-related information. To eliminate the semantic gap between the inputs of two streams, we propose a clothing bias diminishing module specific to the main stream to regularize the features of clothing-relevant regions. Extensive experimental results demonstrate that our framework outperforms other baseline models on several widely-used CC Re-ID datasets.