 |
    |
Unsupervised Video Adaptation for Parsing Human Motion*
Haoquan Shen1, Shoou-I Yu2, Yi Yang3, Deyu Meng4, and Alexander Hauptmann2
1School of Computer Science, Zhejiang University, China
shenhaoquan@gmail.com
2School of Computer Science, Carnegie Mellon University, USA
iyu@cs.cmu.edu
alex@cs.cmu.edu
3ITEE, The University of Queensland, Australia
yee.i.yang@gmail.com
4School of Mathematics and Statistics, Xi’an Jiaotong University, China
dymeng@mail.xjtu.edu.cn
Abstract. In this paper, we propose a method to parse human motion in unconstrained Internet videos without labeling any videos for training. We use the training samples from a public image pose dataset to avoid the tediousness of labeling video streams. There are two main problems confronted. First, the distribution of images and videos are different. Second, no temporal information is available in the training images. To smooth the inconsistency between the labeled images and unlabeled videos, our algorithm iteratively incorporates the pose knowledge harvested from the testing videos into the image pose detector via an adjust-and-refine method. During this process, continuity and tracking constraints are imposed to leverage the spatio-temporal information only available in videos. For our experiments, we have collected two datasets from YouTube and experiments show that our method achieves good performance for parsing human motions. Furthermore, we found that our method achieves better performance by using unlabeled video than adding more labeled pose images into the training set.
Keywords: Unsupervised Video Pose Estimation, Image to Video Adaptation, Unconstrained Internet Videos
Electronic Supplementary Material:
LNCS 8693, p. 347 ff.
lncs@springer.com
© Springer International Publishing Switzerland 2014