InsertAnywhere: Bridging 4D Scene Geometry and Diffusion Models for Realistic Video Object Insertion

TL;DR

InsertAnywhere is a novel framework that combines 4D scene understanding with diffusion models to achieve realistic, geometrically consistent video object insertion with proper occlusion and lighting effects.

Contribution

The paper introduces a 4D aware mask generation module and a diffusion-based video synthesis extension, along with the ROSE++ dataset, advancing realistic video object insertion techniques.

Findings

Outperforms existing models in realism and coherence

Achieves geometrically plausible object insertions

Handles occlusion and lighting effects effectively

Abstract

Recent advances in diffusion-based video generation have opened new possibilities for controllable video editing, yet realistic video object insertion (VOI) remains challenging due to limited 4D scene understanding and inadequate handling of occlusion and lighting effects. We present InsertAnywhere, a new VOI framework that achieves geometrically consistent object placement and appearance-faithful video synthesis. Our method begins with a 4D aware mask generation module that reconstructs the scene geometry and propagates user specified object placement across frames while maintaining temporal coherence and occlusion consistency. Building upon this spatial foundation, we extend a diffusion based video generation model to jointly synthesize the inserted object and its surrounding local variations such as illumination and shading. To enable supervised training, we introduce ROSE++, an illumination aware synthetic dataset constructed by transforming the ROSE object removal dataset into triplets of object removed video, object present video, and a VLM generated reference image. Through extensive experiments, we demonstrate that our framework produces geometrically plausible and visually coherent object insertions across diverse real world scenarios, significantly outperforming existing research and commercial models.