metadata
license: mit
language:
- en
pretty_name: VSCrowd
size_categories:
- 10K<n<100K
tags:
- computer vision
The VSCrowd dataset proposed in "Video Crowd Localization with Multi-focus Gaussian Neighborhood Attention and a Large-Scale Benchmark".
Please find the usage at this GitHub Page.