PLOT: Text-based Person Search with Part Slot Attention for Corresponding Part Discovery
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917780290273280 |
|---|---|
| author | Park, Jicheol Kim, Dongwon Jeong, Boseung Kwak, Suha |
| author_facet | Park, Jicheol Kim, Dongwon Jeong, Boseung Kwak, Suha |
| contents | Text-based person search, employing free-form text queries to identify individuals within a vast image collection, presents a unique challenge in aligning visual and textual representations, particularly at the human part level. Existing methods often struggle with part feature extraction and alignment due to the lack of direct part-level supervision and reliance on heuristic features. We propose a novel framework that leverages a part discovery module based on slot attention to autonomously identify and align distinctive parts across modalities, enhancing interpretability and retrieval accuracy without explicit part-level correspondence supervision. Additionally, text-based dynamic part attention adjusts the importance of each part, further improving retrieval outcomes. Our method is evaluated on three public benchmarks, significantly outperforming existing methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_13475 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | PLOT: Text-based Person Search with Part Slot Attention for Corresponding Part Discovery Park, Jicheol Kim, Dongwon Jeong, Boseung Kwak, Suha Computer Vision and Pattern Recognition Text-based person search, employing free-form text queries to identify individuals within a vast image collection, presents a unique challenge in aligning visual and textual representations, particularly at the human part level. Existing methods often struggle with part feature extraction and alignment due to the lack of direct part-level supervision and reliance on heuristic features. We propose a novel framework that leverages a part discovery module based on slot attention to autonomously identify and align distinctive parts across modalities, enhancing interpretability and retrieval accuracy without explicit part-level correspondence supervision. Additionally, text-based dynamic part attention adjusts the importance of each part, further improving retrieval outcomes. Our method is evaluated on three public benchmarks, significantly outperforming existing methods. |
| title | PLOT: Text-based Person Search with Part Slot Attention for Corresponding Part Discovery |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2409.13475 |