From Global to Local: Rethinking CLIP Feature Aggregation for Person Re-Identification
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908990814814208 |
|---|---|
| author | Zheng, Aotian Sun, Winston Alattar, Bahaa Ablavsky, Vitaly Hwang, Jenq-Neng |
| author_facet | Zheng, Aotian Sun, Winston Alattar, Bahaa Ablavsky, Vitaly Hwang, Jenq-Neng |
| contents | CLIP-based person re-identification (ReID) methods aggregate spatial features into a single global \texttt{[CLS]} token optimized for image-text alignment rather than spatial selectivity, making representations fragile under occlusion and cross-camera variation. We propose SAGA-ReID, which reconstructs identity representations by aligning intermediate patch tokens with anchor vectors parameterized in CLIP's text embedding space -- emphasizing spatially stable evidence while suppressing corrupted or absent regions, without requiring textual descriptions of individual images. Controlled experiments isolate the aggregation mechanism under two qualitatively distinct conditions -- synthetic masking, where identity signal is absent, and realistic human distractors, where an overlapping person introduces semantically confusing signal -- with SAGA's advantage over global pooling growing substantially as occlusion increases across both conditions. Benchmark evaluations confirm consistent gains over CLIP-ReID across standard and occluded settings, with the largest improvements where global pooling is most unreliable: up to +10.6 Rank-1 on occluded benchmarks. SAGA's aggregation outperforms dedicated sequential patch aggregation on a stronger backbone, confirming that structured reconstruction addresses a bottleneck that backbone quality and architectural complexity alone cannot resolve. Code available at https://github.com/ipl-uw/Structured-Anchor-Guided-Aggregation-for-ReID. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_22190 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | From Global to Local: Rethinking CLIP Feature Aggregation for Person Re-Identification Zheng, Aotian Sun, Winston Alattar, Bahaa Ablavsky, Vitaly Hwang, Jenq-Neng Computer Vision and Pattern Recognition Artificial Intelligence CLIP-based person re-identification (ReID) methods aggregate spatial features into a single global \texttt{[CLS]} token optimized for image-text alignment rather than spatial selectivity, making representations fragile under occlusion and cross-camera variation. We propose SAGA-ReID, which reconstructs identity representations by aligning intermediate patch tokens with anchor vectors parameterized in CLIP's text embedding space -- emphasizing spatially stable evidence while suppressing corrupted or absent regions, without requiring textual descriptions of individual images. Controlled experiments isolate the aggregation mechanism under two qualitatively distinct conditions -- synthetic masking, where identity signal is absent, and realistic human distractors, where an overlapping person introduces semantically confusing signal -- with SAGA's advantage over global pooling growing substantially as occlusion increases across both conditions. Benchmark evaluations confirm consistent gains over CLIP-ReID across standard and occluded settings, with the largest improvements where global pooling is most unreliable: up to +10.6 Rank-1 on occluded benchmarks. SAGA's aggregation outperforms dedicated sequential patch aggregation on a stronger backbone, confirming that structured reconstruction addresses a bottleneck that backbone quality and architectural complexity alone cannot resolve. Code available at https://github.com/ipl-uw/Structured-Anchor-Guided-Aggregation-for-ReID. |
| title | From Global to Local: Rethinking CLIP Feature Aggregation for Person Re-Identification |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2604.22190 |