computer-vision-1
Long-video Activity Recognition
3260 papers • 126 benchmarks • 313 datasets
Long-video Activity Recognition (LAR) focuses on modeling long-term relations among all actions in a long video. LAR aims to recognize all actions within each long video, under the weak supervision of the video-level action category set. The mean average precision metric (mAP) is used for evaluation.
(Image credit: Papersgraph)
Benchmarks
These leaderboards are used to track progress in long-video-activity-recognition-1
Trend
Dataset
Best Model
Actions
Breakfast
Breakfast
Libraries
iUse these libraries to find long-video-activity-recognition-1 models and implementations
Datasets
Subtasks
No subtasks available.
Most implemented papers
Timeception for Complex Action Recognition
E. Gavves, A. Smeulders, Noureldien Hussein•Mon Dec 03 2018
Timeception achieves impressive accuracy in recognizing the human activities of Charades, Breakfast Actions and MultiTHUMOS, and it is demonstrated that Timeception learns long-range temporal dependencies and tolerate temporal extents of complex actions.
230