Weakly supervised 2D human pose transfer
| Autor: | Liu Yajie, Hui Huang, Dani Lischinski, Zhizhao Lin, Qian Zheng, Daniel Cohen-Or |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
General Computer Science
Artificial neural network Computer science business.industry Transfer (computing) ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Pattern recognition Artificial intelligence Skeleton (category theory) business ENCODE ComputingMethodologies_COMPUTERGRAPHICS Task (project management) |
| Zdroj: | Science China Information Sciences. 64 |
| ISSN: | 1869-1919 1674-733X |
| DOI: | 10.1007/s11432-021-3301-5 |
| Popis: | We present a novel method for pose transfer between two 2D human skeletons. When the bone lengths and proportions between the two skeletons are significantly different, pose transfer becomes a challenging task, which cannot be accomplished by simply copying the joint positions or the bone directions. Our data-driven approach utilizes a deep neural network trained, in a weakly supervised fashion, to encode a skeleton into two separate latent codes, one representing its pose, and another representing the skeleton’s proportions (skeleton-ID). The network is given two skeletons, and learns to combine the pose of one with the skeleton-ID of the other. Lacking supervision on the poses, we develop a novel loss that qualitatively compares poses of different skeletons. We evaluate the performance of our method on a large set of poses. The advantages of avoiding supervision are demonstrated by showing transfer of extreme poses, as well as between uncommon skeleton proportions. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink