Video-ColBERT: Contextualized Late Interaction for Text-to-Video Retrieval
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909644803276800 |
|---|---|
| author | Reddy, Arun Martin, Alexander Yang, Eugene Yates, Andrew Sanders, Kate Murray, Kenton Kriz, Reno de Melo, Celso M. Van Durme, Benjamin Chellappa, Rama |
| author_facet | Reddy, Arun Martin, Alexander Yang, Eugene Yates, Andrew Sanders, Kate Murray, Kenton Kriz, Reno de Melo, Celso M. Van Durme, Benjamin Chellappa, Rama |
| contents | In this work, we tackle the problem of text-to-video retrieval (T2VR). Inspired by the success of late interaction techniques in text-document, text-image, and text-video retrieval, our approach, Video-ColBERT, introduces a simple and efficient mechanism for fine-grained similarity assessment between queries and videos. Video-ColBERT is built upon 3 main components: a fine-grained spatial and temporal token-wise interaction, query and visual expansions, and a dual sigmoid loss during training. We find that this interaction and training paradigm leads to strong individual, yet compatible, representations for encoding video content. These representations lead to increases in performance on common text-to-video retrieval benchmarks compared to other bi-encoder methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_19009 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Video-ColBERT: Contextualized Late Interaction for Text-to-Video Retrieval Reddy, Arun Martin, Alexander Yang, Eugene Yates, Andrew Sanders, Kate Murray, Kenton Kriz, Reno de Melo, Celso M. Van Durme, Benjamin Chellappa, Rama Computer Vision and Pattern Recognition Information Retrieval In this work, we tackle the problem of text-to-video retrieval (T2VR). Inspired by the success of late interaction techniques in text-document, text-image, and text-video retrieval, our approach, Video-ColBERT, introduces a simple and efficient mechanism for fine-grained similarity assessment between queries and videos. Video-ColBERT is built upon 3 main components: a fine-grained spatial and temporal token-wise interaction, query and visual expansions, and a dual sigmoid loss during training. We find that this interaction and training paradigm leads to strong individual, yet compatible, representations for encoding video content. These representations lead to increases in performance on common text-to-video retrieval benchmarks compared to other bi-encoder methods. |
| title | Video-ColBERT: Contextualized Late Interaction for Text-to-Video Retrieval |
| topic | Computer Vision and Pattern Recognition Information Retrieval |
| url | https://arxiv.org/abs/2503.19009 |