The ZJU-MoCap dataset is a prominent benchmark in human modeling from videos. The dataset includes 6 human subjects, each with 100 frames collected from one camera for training and the remaining cameras for evaluation.
BibTex:
Before browse our site, please accept our cookies policy
