Clothes-Changing Person Re-identification Based On Skeleton Dynamics
This addresses the challenge of recognizing individuals across videos despite clothing changes, which is important for surveillance and security applications, and is incremental as it builds on existing skeleton-based approaches.
The paper tackles the problem of clothes-changing person re-identification by proposing a method that uses only skeleton data, avoiding reliance on appearance features, and achieves state-of-the-art performance on the CCVID dataset.
Clothes-Changing Person Re-Identification (ReID) aims to recognize the same individual across different videos captured at various times and locations. This task is particularly challenging due to changes in appearance, such as clothing, hairstyle, and accessories. We propose a Clothes-Changing ReID method that uses only skeleton data and does not use appearance features. Traditional ReID methods often depend on appearance features, leading to decreased accuracy when clothing changes. Our approach utilizes a spatio-temporal Graph Convolution Network (GCN) encoder to generate a skeleton-based descriptor for each individual. During testing, we improve accuracy by aggregating predictions from multiple segments of a video clip. Evaluated on the CCVID dataset with several different pose estimation models, our method achieves state-of-the-art performance, offering a robust and efficient solution for Clothes-Changing ReID.