# Cross-Identity Motion Transfer for Arbitrary Objects through Pose-Attentive Video Reassembling

arXiv: https://arxiv.org/abs/2007.08786 · HTML: https://mlsanity.com/paper/2007.08786/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Jul 2020.

## Built on by (most cited first, 9 of 9)

- [View Generalization for Single Image Textured 3D Models](https://mlsanity.com/paper/2106.06533/) (2021)
- [Fine Detailed Texture Learning for 3D Meshes with Generative Models](https://mlsanity.com/paper/2203.09362/) (2022)
- [Cross-identity Video Motion Retargeting with Joint Transformation and Synthesis](https://mlsanity.com/paper/2210.01559/) (2022)
- [Motion and Appearance Adaptation for Cross-Domain Motion Transfer](https://mlsanity.com/paper/2209.14529/) (2022)
- [REMOT: A Region-to-Whole Framework for Realistic Human Motion Transfer](https://mlsanity.com/paper/2209.00475/) (2022)
- [IF-MDM: Implicit Face Motion Diffusion Model for High-Fidelity Realtime Talking Head Generation](https://mlsanity.com/paper/2412.04000/) (2024)
- [PriorityCut: Occlusion-guided Regularization for Warp-based Image Animation](https://mlsanity.com/paper/2103.11600/) (2021)
- [OOOE: Only-One-Object-Exists Assumption to Find Very Small Objects in Chest Radiographs](https://mlsanity.com/paper/2210.06806/) (2022)
- [DISPLAY: Directable Human-Object Interaction Video Generation via Sparse Motion Guidance and Multi-Task Auxiliary](https://mlsanity.com/paper/2603.09883/) (2026)

## Abstract (arXiv)

We propose an attention-based networks for transferring motions between arbitrary objects. Given a source image(s) and a driving video, our networks animate the subject in the source images according to the motion in the driving video. In our attention mechanism, dense similarities between the learned keypoints in the source and the driving images are computed in order to retrieve the appearance information from the source images. Taking a different approach from the well-studied warping based models, our attention-based model has several advantages. By reassembling non-locally searched pieces from the source contents, our approach can produce more realistic outputs. Furthermore, our system can make use of multiple observations of the source appearance (e.g. front and sides of faces) to make the results more accurate. To reduce the training-testing discrepancy of the self-supervised learning, a novel cross-identity training scheme is additionally introduced. With the training scheme, our networks is trained to transfer motions between different subjects, as in the real testing scenario. Experimental results validate that our method produces visually pleasing results in various object domains, showing better performances compared to previous works.