Journal Article10.1007/s11042-024-18238-4
Contrastive disentanglement for self-supervised motion style transfer
Zizhao Wu,Siyuan Mao,Cheng Zhang,Yigang Wang,Ming Zeng +4 more
About: This article is published in Multimedia Tools and Applications. The article was published on 30 Jan 2024.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
References
Realtime style transfer for unlabeled heterogeneous human motion
Shihong Xia,Congyi Wang,Jinxiang Chai,Jessica K. Hodgins +3 more
- 27 Jul 2015
TL;DR: A novel solution for realtime generation of stylistic human motion that automatically transforms unlabeled, heterogeneous motion data into new styles and introduces an efficient local regression model to predict the timings of synthesized poses in the output style.
218
Exploring Disentangled Feature Representation Beyond Face Identification
Yu Liu,Fangyin Wei,Jing Shao,Lu Sheng,Junjie Yan,Xiaogang Wang +5 more
- 10 Apr 2018
TL;DR: Comprehensive evaluations demonstrate that the proposed features not only preserve state-of-the-art identity verification performance on LFW, but also acquire comparable discriminative power for face attribute recognition on CelebA and LFWA.
208
Local motion phases for learning multi-contact character movements
TL;DR: A novel framework to learn fast and dynamic character interactions that involve multiple contacts between the body and an object, another character and the environment, from a rich, unstructured motion capture database is proposed.
Domain Enhanced Arbitrary Image Style Transfer via Contrastive Learning
Yuxin Zhang,Fan Tang,Weiming Dong,Haibin Huang,Chongyang Ma,Tong-Yee Lee,Changsheng Xu +6 more
- 19 May 2022
TL;DR: This work presents Contrastive Arbitrary Style Transfer (CAST), which is a new style representation learning and style transfer method via contrastive learning that achieves significantly better results compared to those obtained via state-of-the-art methods.
Unpaired Motion Style Transfer from Video to Animation
TL;DR: This paper presents a novel data-driven framework for motion style transfer, which learns from an unpaired collection of motions with style labels, and enables transferring motion styles not observed during training, and is the first to demonstrate style transfer directly from videos to 3D animations.