5 datasets found

Generating human action proposals in untrimmed videos is an important yet challenging task with wide applications. Current methods often suffer from the noisy boundary locations...

The dataset used in the paper is ActivityNet, MSR-VTT, and MSVD. The authors used these datasets for text-to-video retrieval tasks.

Weakly-Supervised Temporal Action Localization (WSTAL) aims to localize actions in untrimmed videos with only video-level labels.

The ActivityNet Captions is a benchmark dataset proposed for dense video captioning. There are 20K untrimmed videos in total, and each video has several annotated segments with...

Temporal activity detection has drawn increasing interests in both academic and industry communities due to its vast potential applications in security surveillance, behavior...