metadata
task_categories:
- visual-question-answering
language:
- en
size_categories:
- 1K<n<10K
license: mit
Towards Event-oriented Long Video Understanding
๐ Overview
We introduce Event-Bench, an event-oriented long video understanding benchmark built on existing datasets and human annotations. Event-Bench consists of three event understanding abilities and six event-related tasks, including 2,190 test instances to comprehensively evaluate the ability to understand video events.
Event-Bench provides a systematic comparison across different kinds of capabilities for existing video MLLMs, and points out the major shortcomings of open-source MLLMs.
๐ Dataset
Download the raw videos in VNBench from the google drive link.
License:
Event-Bench is only used for academic research. Commercial use in any form is prohibited.
๐ฎ Evaluation Pipeline
Please refer to https://github.com/RUCAIBox/Event-Bench
๐ Experimental Results
- Evaluation results of different Video MLLMs.
Citation
If you find our work helpful for your research, please consider citing our work.