Demo for Martial Arts, Dancing and Sports Dataset

There are videos for pose tracking results. There are 30 videos showing multi-view results (6 for each action category), and another  30 videos showing results on depth video (6 for each action category). The multi-view videos present the original image, the ground-truth pose and results of baseline algorithms.

The baselines for multi-view are the bi-directional likelihood, the robust ECPBL likelihood and the TGP algorithm. The tracking error is showed in the top-right corner for each algorithm. A green value indicates that the error is smaller than 80 mm, while red one indicates that the error is larger than 80mm. The depth videos show the original right image, the depth map, the ground-truth skeleton with results of baseline algorithms. For the Tai-chi, the baselines are the basic linear likelihood, uni-directional likelihood and the robust likelihood with APF tracker, as well as the results of the PDT tracker, the TGP algorithm and the GMM-based shape and pose estimation algorithm. For other action categories, the baselines are the linear likelihood, uni-directional likelihood, the robust likelihood and the TGP algorithm.

Youtube playlists

Demo videos can be viewed on Youtube.

You can download the videos directly below.

Multiview demo videos

Depth demo videos


Back to project page