Martial Arts, Dancing and Sports dataset: A challenging stereo and multi-view dataset for 3D human pose estimation

作者：

Highlights：

• We propose a new dataset called the Martial Arts, Dancing and Sports dataset for 3D human pose estimation.

• The dataset contains challenging actions from Tai-chi, Karate, jazz, hip-hop and sports.

• It contains 30 multi-view videos and 30 stereo depth videos, with a total of 53,000 frames.

• We provide initial results using several baseline algorithms.

摘要

•We propose a new dataset called the Martial Arts, Dancing and Sports dataset for 3D human pose estimation.•The dataset contains challenging actions from Tai-chi, Karate, jazz, hip-hop and sports.•It contains 30 multi-view videos and 30 stereo depth videos, with a total of 53,000 frames.•We provide initial results using several baseline algorithms.

论文关键词：Human pose estimation,Robust tracking,Evaluation,Martial arts,Dancing and sports

论文评审过程：Received 27 September 2015, Revised 13 August 2016, Accepted 11 February 2017, Available online 21 February 2017, Version of Record 11 March 2017.

论文官网地址：https://doi.org/10.1016/j.imavis.2017.02.002