Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.21094 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908731186348032 |
|---|---|
| author | Cao, Zhe Wang, Tao Wang, Jiaming Wang, Yanghai Zhang, Yuanxing Chen, Jialu Deng, Miao Wang, Jiahao Guo, Yubin Liao, Chenxi Zhang, Yize Zhang, Zhaoxiang Liu, Jiaheng |
| author_facet | Cao, Zhe Wang, Tao Wang, Jiaming Wang, Yanghai Zhang, Yuanxing Chen, Jialu Deng, Miao Wang, Jiahao Guo, Yubin Liao, Chenxi Zhang, Yize Zhang, Zhaoxiang Liu, Jiaheng |
| contents | Text-to-Audio-Video (T2AV) generation aims to synthesize temporally coherent video and semantically synchronized audio from natural language, yet its evaluation remains fragmented, often relying on unimodal metrics or narrowly scoped benchmarks that fail to capture cross-modal alignment, instruction following, and perceptual realism under complex prompts. To address this limitation, we present T2AV-Compass, a unified benchmark for comprehensive evaluation of T2AV systems, consisting of 500 diverse and complex prompts constructed via a taxonomy-driven pipeline to ensure semantic richness and physical plausibility. Besides, T2AV-Compass introduces a dual-level evaluation framework that integrates objective signal-level metrics for video quality, audio quality, and cross-modal alignment with a subjective MLLM-as-a-Judge protocol for instruction following and realism assessment. Extensive evaluation of 11 representative T2AVsystems reveals that even the strongest models fall substantially short of human-level realism and cross-modal consistency, with persistent failures in audio realism, fine-grained synchronization, instruction following, etc. These results indicate significant improvement room for future models and highlight the value of T2AV-Compass as a challenging and diagnostic testbed for advancing text-to-audio-video generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_21094 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | T2AV-Compass: Towards Unified Evaluation for Text-to-Audio-Video Generation Cao, Zhe Wang, Tao Wang, Jiaming Wang, Yanghai Zhang, Yuanxing Chen, Jialu Deng, Miao Wang, Jiahao Guo, Yubin Liao, Chenxi Zhang, Yize Zhang, Zhaoxiang Liu, Jiaheng Computer Vision and Pattern Recognition Text-to-Audio-Video (T2AV) generation aims to synthesize temporally coherent video and semantically synchronized audio from natural language, yet its evaluation remains fragmented, often relying on unimodal metrics or narrowly scoped benchmarks that fail to capture cross-modal alignment, instruction following, and perceptual realism under complex prompts. To address this limitation, we present T2AV-Compass, a unified benchmark for comprehensive evaluation of T2AV systems, consisting of 500 diverse and complex prompts constructed via a taxonomy-driven pipeline to ensure semantic richness and physical plausibility. Besides, T2AV-Compass introduces a dual-level evaluation framework that integrates objective signal-level metrics for video quality, audio quality, and cross-modal alignment with a subjective MLLM-as-a-Judge protocol for instruction following and realism assessment. Extensive evaluation of 11 representative T2AVsystems reveals that even the strongest models fall substantially short of human-level realism and cross-modal consistency, with persistent failures in audio realism, fine-grained synchronization, instruction following, etc. These results indicate significant improvement room for future models and highlight the value of T2AV-Compass as a challenging and diagnostic testbed for advancing text-to-audio-video generation. |
| title | T2AV-Compass: Towards Unified Evaluation for Text-to-Audio-Video Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.21094 |