Generating High-Fidelity Clothes-Varying Samples and Optimizing Conflicting Objectives for Clothes-Changing Person Re-Identification
Clothes-changing person re-identification (CC-ReID) aims to retrieve images of the same person wearing different outfits. This work uncovers the conflicting relationship between standard and clothes-changing learning objectives in CC-ReID, and proposes to generate high-fidelity clothes-varying synthetic data and formulate CC-ReID as a multi-objective optimization problem to mitigate the conflicts.
Clothes-Changing Person Re-Identification with Feasibility-Aware Intermediary Matching
The core message of this paper is to propose a Feasibility-Aware Intermediary Matching (FAIM) framework that utilizes both clothes-relevant and clothes-irrelevant features to perform intermediary-assisted matching for clothes-changing person re-identification, while also assessing the feasibility of the intermediary matching process.
© 2024 by Linnk AI