VIDEOCOUNT
收藏资源简介:
VIDEOCOUNT是一个视频计数数据集,旨在评估开放世界视频对象计数任务的算法性能。该数据集由三个基准测试组成,涵盖了141个对象类别和每个视频1到1203个对象的不同对象计数。数据集的构建使用了TAO和MOT20跟踪数据集,以及从自然栖息地监测企鹅和金属合金结晶过程的X射线视频中捕获的视频。VIDEOCOUNT数据集可用于评估COUNTVID等模型在开放世界视频对象计数任务中的性能,旨在解决视频中的对象计数问题,尤其是在拥挤场景中避免重复计数和识别重新出现的对象。
VIDEOCOUNT is a video counting dataset developed to evaluate the performance of algorithms for open-world video object counting tasks. This dataset consists of three benchmark tests, covering 141 object categories and varying object counts ranging from 1 to 1203 objects per video. The dataset is constructed using the TAO and MOT20 tracking datasets, alongside videos extracted from X-ray footage used for monitoring penguins in natural habitats and capturing the crystallization process of metal alloys. The VIDEOCOUNT dataset can be utilized to assess the performance of models such as COUNTVID in open-world video object counting tasks, with the objective of addressing video object counting challenges, especially avoiding duplicate counts and recognizing re-occurring objects in crowded scenarios.
- 1Open-World Object Counting in Videos牛津大学视觉几何组 · 2025年



