人类行动数据集（视频）

UCF YOUTUBE ACTION DATASET ======================================================================================================== 1. It contains 11 action categories: basketball shooting, biking/cycling, diving, golf swinging, horseback riding, soccer juggling, swinging, tennis swinging, trampoline jumping, volleyball spiking, and walking with a dog 2. For each category, the videos are grouped into 25 groups each with more than 4 action clips. The video clips in the same group may share some common characteristics, such as the same subject, similar background, similar viewpoint, etc. 3. The videos are in ms mpeg4 format. You need to install the right video codec (e.g. K-lite Codec Pack contains a cellection of Codecs) to access them. 4. Annotions are avilable in each folder in VIPER-GT XML format. 5. If you need a reference for this dataset, please refer to paper: J. Liu, J. Luo and M. Shah, Recognizing realistic actions from videos "in the wild", CVPR 2009, Miami, FL. (In this study, all the videos are selected for action biking and walking; for the rest of action classes, we only used the first four video clips (01 to 04) if any group contains more than four video clips). DISCLAIMER ===================================================================================================== The dataset is intended only for non-commercial research purposes. Please consult the terms and conditions of use of these videos from the YouTube website.