VOID: Video Object and Interaction Deletion

TL;DR

VOID is a novel video object removal framework that achieves physically plausible inpainting in complex scenarios involving object interactions by combining scene understanding and diffusion models.

Contribution

The paper introduces a new dataset and a method that integrates vision-language models with video diffusion to improve scene consistency after object removal.

Findings

Outperforms prior methods in preserving scene dynamics.

Generates more physically consistent inpainted videos.

Effective on both synthetic and real data.

Abstract

Existing video object removal methods excel at inpainting content "behind" the object and correcting appearance-level artifacts such as shadows and reflections. However, when the removed object has more significant interactions, such as collisions with other objects, current models fail to correct them and produce implausible results. We present VOID, a video object removal framework designed to perform physically-plausible inpainting in these complex scenarios. To train the model, we generate a new paired dataset of counterfactual object removals using Kubric and HUMOTO, where removing an object requires altering downstream physical interactions. During inference, a vision-language model identifies regions of the scene affected by the removed object. These regions are then used to guide a video diffusion model that generates physically consistent counterfactual outcomes. Experiments on both synthetic and real data show that our approach better preserves consistent scene dynamics after object removal compared to prior video object removal methods. We hope this framework sheds light on how to make video editing models better simulators of the world through high-level causal reasoning.