AutoAD-Zero: A Training-Free Framework for Zero-Shot Audio Description
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866929600051806208 |
|---|---|
| author | Xie, Junyu Han, Tengda Bain, Max Nagrani, Arsha Varol, Gül Xie, Weidi Zisserman, Andrew |
| author_facet | Xie, Junyu Han, Tengda Bain, Max Nagrani, Arsha Varol, Gül Xie, Weidi Zisserman, Andrew |
| contents | Our objective is to generate Audio Descriptions (ADs) for both movies and TV series in a training-free manner. We use the power of off-the-shelf Visual-Language Models (VLMs) and Large Language Models (LLMs), and develop visual and text prompting strategies for this task. Our contributions are three-fold: (i) We demonstrate that a VLM can successfully name and refer to characters if directly prompted with character information through visual indications without requiring any fine-tuning; (ii) A two-stage process is developed to generate ADs, with the first stage asking the VLM to comprehensively describe the video, followed by a second stage utilising a LLM to summarise dense textual information into one succinct AD sentence; (iii) A new dataset for TV audio description is formulated. Our approach, named AutoAD-Zero, demonstrates outstanding performance (even competitive with some models fine-tuned on ground truth ADs) in AD generation for both movies and TV series, achieving state-of-the-art CRITIC scores. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_15850 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | AutoAD-Zero: A Training-Free Framework for Zero-Shot Audio Description Xie, Junyu Han, Tengda Bain, Max Nagrani, Arsha Varol, Gül Xie, Weidi Zisserman, Andrew Computer Vision and Pattern Recognition Our objective is to generate Audio Descriptions (ADs) for both movies and TV series in a training-free manner. We use the power of off-the-shelf Visual-Language Models (VLMs) and Large Language Models (LLMs), and develop visual and text prompting strategies for this task. Our contributions are three-fold: (i) We demonstrate that a VLM can successfully name and refer to characters if directly prompted with character information through visual indications without requiring any fine-tuning; (ii) A two-stage process is developed to generate ADs, with the first stage asking the VLM to comprehensively describe the video, followed by a second stage utilising a LLM to summarise dense textual information into one succinct AD sentence; (iii) A new dataset for TV audio description is formulated. Our approach, named AutoAD-Zero, demonstrates outstanding performance (even competitive with some models fine-tuned on ground truth ADs) in AD generation for both movies and TV series, achieving state-of-the-art CRITIC scores. |
| title | AutoAD-Zero: A Training-Free Framework for Zero-Shot Audio Description |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2407.15850 |