TUNA: Comprehensive Fine-grained Temporal Understanding Evaluation on Dense Dynamic Videos
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909641404841984 |
|---|---|
| author | Kong, Fanheng Zhang, Jingyuan Zhang, Hongzhi Feng, Shi Wang, Daling Yu, Linhao Ji, Xingguang Tian, Yu W., Victoria Zhang, Fuzheng |
| author_facet | Kong, Fanheng Zhang, Jingyuan Zhang, Hongzhi Feng, Shi Wang, Daling Yu, Linhao Ji, Xingguang Tian, Yu W., Victoria Zhang, Fuzheng |
| contents | Videos are unique in their integration of temporal elements, including camera, scene, action, and attribute, along with their dynamic relationships over time. However, existing benchmarks for video understanding often treat these properties separately or narrowly focus on specific aspects, overlooking the holistic nature of video content. To address this, we introduce TUNA, a temporal-oriented benchmark for fine-grained understanding on dense dynamic videos, with two complementary tasks: captioning and QA. Our TUNA features diverse video scenarios and dynamics, assisted by interpretable and robust evaluation criteria. We evaluate several leading models on our benchmark, providing fine-grained performance assessments across various dimensions. This evaluation reveals key challenges in video temporal understanding, such as limited action description, inadequate multi-subject understanding, and insensitivity to camera motion, offering valuable insights for improving video understanding models. The data and code are available at https://friedrichor.github.io/projects/TUNA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_20124 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TUNA: Comprehensive Fine-grained Temporal Understanding Evaluation on Dense Dynamic Videos Kong, Fanheng Zhang, Jingyuan Zhang, Hongzhi Feng, Shi Wang, Daling Yu, Linhao Ji, Xingguang Tian, Yu W., Victoria Zhang, Fuzheng Computer Vision and Pattern Recognition Multimedia Videos are unique in their integration of temporal elements, including camera, scene, action, and attribute, along with their dynamic relationships over time. However, existing benchmarks for video understanding often treat these properties separately or narrowly focus on specific aspects, overlooking the holistic nature of video content. To address this, we introduce TUNA, a temporal-oriented benchmark for fine-grained understanding on dense dynamic videos, with two complementary tasks: captioning and QA. Our TUNA features diverse video scenarios and dynamics, assisted by interpretable and robust evaluation criteria. We evaluate several leading models on our benchmark, providing fine-grained performance assessments across various dimensions. This evaluation reveals key challenges in video temporal understanding, such as limited action description, inadequate multi-subject understanding, and insensitivity to camera motion, offering valuable insights for improving video understanding models. The data and code are available at https://friedrichor.github.io/projects/TUNA. |
| title | TUNA: Comprehensive Fine-grained Temporal Understanding Evaluation on Dense Dynamic Videos |
| topic | Computer Vision and Pattern Recognition Multimedia |
| url | https://arxiv.org/abs/2505.20124 |