AutoAD III: The Prequel -- Back to the Pixels
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_ | 1866909177625968640 |
|---|---|
| author | Han, Tengda Bain, Max Nagrani, Arsha Varol, Gül Xie, Weidi Zisserman, Andrew |
| author_facet | Han, Tengda Bain, Max Nagrani, Arsha Varol, Gül Xie, Weidi Zisserman, Andrew |
| contents | Generating Audio Description (AD) for movies is a challenging task that requires fine-grained visual understanding and an awareness of the characters and their names. Currently, visual language models for AD generation are limited by a lack of suitable training data, and also their evaluation is hampered by using performance measures not specialized to the AD domain. In this paper, we make three contributions: (i) We propose two approaches for constructing AD datasets with aligned video data, and build training and evaluation datasets using these. These datasets will be publicly released; (ii) We develop a Q-former-based architecture which ingests raw video and generates AD, using frozen pre-trained visual encoders and large language models; and (iii) We provide new evaluation metrics to benchmark AD quality that are well-matched to human performance. Taken together, we improve the state of the art on AD generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_14412 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | AutoAD III: The Prequel -- Back to the Pixels Han, Tengda Bain, Max Nagrani, Arsha Varol, Gül Xie, Weidi Zisserman, Andrew Computer Vision and Pattern Recognition Generating Audio Description (AD) for movies is a challenging task that requires fine-grained visual understanding and an awareness of the characters and their names. Currently, visual language models for AD generation are limited by a lack of suitable training data, and also their evaluation is hampered by using performance measures not specialized to the AD domain. In this paper, we make three contributions: (i) We propose two approaches for constructing AD datasets with aligned video data, and build training and evaluation datasets using these. These datasets will be publicly released; (ii) We develop a Q-former-based architecture which ingests raw video and generates AD, using frozen pre-trained visual encoders and large language models; and (iii) We provide new evaluation metrics to benchmark AD quality that are well-matched to human performance. Taken together, we improve the state of the art on AD generation. |
| title | AutoAD III: The Prequel -- Back to the Pixels |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2404.14412 |