Transfer4D: A Framework for Frugal Motion Capture and Deformation Transfer
Shubh Maheshwari, Rahul Narain, Ramya Hebbalaguppe
Abstract
Figure 1 . Transfer4D is an unsupervised technique for intra-category motion transfer from monocular depth videos (highlighted in the green box) to virtual characters of varying shapes (blue). It is category-agnostic, permitting motion transfer between quadrupeds (left), between bipeds (center and right), and in general between any characters with similar topology. Skeletons shown in yellow are extracted using our proposed unsupervised skeletonization algorithm. Please refer to the demo video on the project webpage: https: //transfer4d.github.io/ and Fig. 3 for diverse animation sequences ranging bipeds and quadrupeds.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Your agent calls
Luneget_paper_fulltext
Free to start. No credit card required.
Terminal
Install the CLIlune papers fulltext 1288484c-ceb6-43fa-8248-fd06a30b0148Cited by top-tier papers2
- OmniMotionGPT: Animal Motion Generation with Limited DataZhangsihao Yang, Mingyuan Zhou, Mengyi Shan, Bingbing Wen et al.CVPR 2024 · 6 citations
- ARMO: Autoregressive Rigging for Multi-Category ObjectsMingze Sun, Shiwei Mao, Keyi Chen, Yurun Chen et al.ICCV 2025 · 3 citations
Builds on18
- Nerfies: Deformable Neural Radiance FieldsKeunhong Park, Utkarsh Sinha, Jonathan T. Barron, Sofien Bouaziz et al.ICCV 2021 · 1,442 citations
- Non-Rigid Neural Radiance Fields: Reconstruction and Novel View Synthesis of a Dynamic Scene From Monocular VideoEdgar Tretschk, Ayush Tewari, Vladislav Golyanik, Michael Zollhöfer et al.ICCV 2021 · 617 citations
- Skeleton-aware networks for deep motion retargetingKfir Aberman, Peizhuo Li, Dani Lischinski, Olga Sorkine-Hornung et al.SIGGRAPH 2020 · 210 citations
- Unsupervised Deep Learning for Structured Shape MatchingJean-Michel Roufosse, Abhishek Sharma, Maks OvsjanikovICCV 2019 · 160 citations
- 4DComplete: Non-Rigid Motion Estimation Beyond the Observable SurfaceYang Li, Hikari Takehara, Takafumi Taketomi, Bo Zheng et al.ICCV 2021 · 160 citations
Related papers
- Reconstructing Animatable Categories from VideosGengshan Yang, Chaoyang Wang, N. Dinesh Reddy, Deva RamananCVPR 2023
- Motion4Motion: Motion Transfer Across Subjects at InferenceLing-Hao Chen, Zixin Yin, Duomin Wang, Xianfang Zeng et al.SIGGRAPH 2026
- MoCaNet: Motion Retargeting In-the-Wild via Canonicalization NetworksWentao Zhu, Zhuoqian Yang, Ziang Di, Wayne Wu et al.AAAI 2022 · 24 citations
- Behave Your Motion: Habit-preserved Cross-category Animal Motion TransferZhimin Zhang, Bi'an Du, Caoyuan Ma, Zheng Wang et al.ACM MM 2025 · 2 citations
- LASR: Learning Articulated Shape Reconstruction From a Monocular VideoGengshan Yang, Deqing Sun, Varun Jampani, Daniel Vlasic et al.CVPR 2021
