# MV-CoLight: Efficient Object Compositing with Consistent Lighting and Shadow Generation

Source: [MV-CoLight: Efficient Object Compositing with Consistent Lighting and Shadow Generation](<https://www.sota2.com/research/paper/68fa78d11e209131cea7a940>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T04:47:07.371Z

Authors: Kerui Ren, Jiayang Bai, Linning Xu, Lihan Jiang, Jiangmiao Pang, Mulin Yu, Bo Dai

Published: 2025-05-27T17:53:02.000Z

## Abstract

Object compositing offers significant promise for augmented reality (AR) and embodied intelligence applications. Existing approaches predominantly focus on single-image scenarios or intrinsic decomposition techniques, facing challenges with multi-view consistency, complex scenes, and diverse lighting conditions. Recent inverse rendering advancements, such as 3D Gaussian and diffusion-based methods, have enhanced consistency but are limited by scalability, heavy data requirements, or prolonged reconstruction time per scene. To broaden its applicability, we introduce MV-CoLight, a two-stage framework for illumination-consistent object compositing in both 2D images and 3D scenes. Our novel feed-forward architecture models lighting and shadows directly, avoiding the iterative biases of diffusion-based methods. We employ a Hilbert curve-based mapping to align 2D image inputs with 3D Gaussian scene representations seamlessly. To facilitate training and evaluation, we further introduce a large-scale 3D compositing dataset. Experiments demonstrate state-of-the-art harmonized results across standard benchmarks and our dataset, as well as casually captured real-world scenes demonstrate the framework's robustness and wide generalization.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Mesh Insertion on LH3D-Blender](<https://www.sota2.com/research/sota/mesh-insertion-on-lh3d-blender.md>): Mesh Insertion; LH3D-Blender; PSNR: 19.257.

- [Visual Quality Assessment on Mip-NeRF360](<https://www.sota2.com/research/sota/visual-quality-assessment-on-mip-nerf360.md>): Visual Quality Assessment; Mip-NeRF360; Positive Score: 33.9.

- [Object Insertion on LH3D-Ku (test)](<https://www.sota2.com/research/sota/object-insertion-on-lh3d-ku-test.md>): Object Insertion; LH3D-Ku (test); PSNR: 15.99.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2505.21483>)

[PDF](<https://arxiv.org/pdf/2505.21483.pdf>)
