Learning Continual Compatible Representation for Re-indexing Free Lifelong Person Re-identification

Z Cui, J Zhou, X Wang, M Zhu… - Proceedings of the IEEE …, 2024 - openaccess.thecvf.com
Abstract Lifelong Person Re-identification (L-ReID) aims to learn from sequentially collected
data to match a person across different scenes. Once an L-ReID model is updated using …

MixBCT: Towards Self-Adapting Backward-Compatible Training

Y Liang, Y Zhang, S Zhang, Y Wang, S **ao… - arxiv preprint arxiv …, 2023 - arxiv.org
Backward-compatible training circumvents the need for expensive updates to the old gallery
database when deploying an advanced new model in the retrieval system. Previous …

Backward-Compatible Aligned Representations via an Orthogonal Transformation Layer

S Ricci, N Biondi, F Pernici, A Del Bimbo - arxiv preprint arxiv:2408.08793, 2024 - arxiv.org
Visual retrieval systems face significant challenges when updating models with improved
representations due to misalignment between the old and new representations. The costly …

FC-Aligner: A Lightweight Regressor Model for Embedding Space Conversion

ALBV e Silva, RG Ferrari, ÁS Nolibos… - Latinx in AI@ NeurIPS … - openreview.net
In diverse applications like image clustering, facial recognition and text embeddings,
similarity search is critical. Deep models utilize feature embeddings for efficient …