VideoAnydoor: High-fidelity Video Object Insertion with Precise Motion Control

TL;DR

VideoAnydoor is a zero-shot framework for high-fidelity video object insertion that preserves appearance details and offers precise motion control through a pixel warper and a novel training strategy.

Contribution

It introduces a novel pixel warper and training approach enabling detailed appearance preservation and fine-grained motion control in video object insertion.

Findings

Outperforms existing methods in insertion quality

Supports various downstream applications without fine-tuning

Enables precise motion manipulation with key-point trajectories

Abstract

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 weighted 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.