{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"Daily Paper Cast","title":"VideoAnydoor: High-fidelity Video Object Insertion with Precise Motion Control","html":"<iframe width=\"100%\" height=\"180\" frameborder=\"no\" scrolling=\"no\" seamless src=\"https://share.transistor.fm/e/21216081\"></iframe>","width":"100%","height":180,"duration":1155,"description":"\n            🤗 Upvotes: 30 | cs.CV\n\n            Authors:\n            Yuanpeng Tu, Hao Luo, Xi Chen, Sihui Ji, Xiang Bai, Hengshuang Zhao\n\n            Title:\n            VideoAnydoor: High-fidelity Video Object Insertion with Precise Motion Control\n\n            Arxiv:\n            http://arxiv.org/abs/2501.01427v1\n\n            Abstract:\n            Despite significant advancements in video generation, inserting a given object into videos remains a challenging task. The difficulty lies in preserving the appearance details of the reference object and accurately modeling coherent motions at the same time. In this paper, we propose VideoAnydoor, a zero-shot video object insertion framework with high-fidelity detail preservation and precise motion control. Starting from a text-to-video model, we utilize an ID extractor to inject the global identity and leverage a box sequence to control the overall motion. To preserve the detailed appearance and meanwhile support fine-grained motion control, we design a pixel warper. It takes the reference image with arbitrary key-points and the corresponding key-point trajectories as inputs. It warps the pixel details according to the trajectories and fuses the warped features with the diffusion U-Net, thus improving detail preservation and supporting users in manipulating the motion trajectories. In addition, we propose a training strategy involving both videos and static images with a reweight reconstruction loss to enhance insertion quality. VideoAnydoor demonstrates significant superiority over existing methods and naturally supports various downstream applications (e.g., talking head generation, video virtual try-on, multi-region editing) without task-specific fine-tuning.\n            ","thumbnail_url":"https://img.transistorcdn.com/8lOVNnuwhrA3rxrDMv7Osu4j_t1-jORooO6NfGcQhcw/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS81Zjg1/YzRhODczMDU4MmE4/OGMwN2FiNDlmYzI2/MDliMi5qcGVn.webp","thumbnail_width":300,"thumbnail_height":300}