Compression-aware restoration improves 3d reconstruction from multi-view images
CompMVR: Compression-Aware Multi-View Restoration Using Diffusion Models for Geometrically Consistent 3D Reconstruction
Computer Vision and Pattern Recognition
Summary
3D models often rely on photos taken from many angles, but when these photos are compressed to save space, they get blurry and inconsistent across views, making 3D reconstruction harder. The authors found that common compression can cause mismatched features between photos, messing up the 3D modeling process. They created CompMVR, a method that fixes these compressed photos together by considering all views at once to keep them consistent. This improves both the image quality and the accuracy of 3D reconstructions built from those photos.
What this means in practice
- •For 3d graphics developers: Improve 3D model quality by restoring compressed multi-view images with consistent geometric details for better rendering and synthesis.
- •For augmented reality engineers: Enhance AR experiences by refining compressed input images from multiple cameras to provide more accurate spatial mapping and scene understanding.
Authors
Dong-hwi Kim, Chaewon Moon, Hojun Song, Dongbeom Kim, Junyeong Jang, Aro Kim, Gahyeon Kim, Heejung Choi, Jehee Kim, Gianella Cravioto, Sohyun Lee, Gyeongjin Choi, EunHye Jeong, Soo Ye Kim, Jaehyup Lee, Sang-hyo Park
Abstract
Real-world 3D reconstruction pipelines often rely on multi-view images degraded by lossy compression. While compression artifacts have been extensively studied in single-image or video restoration, their impact on multi-view geometry remains largely overlooked. In this work, we identify compression-induced cross-view geometric inconsistency as a critical failure mode, where view-dependent artifacts disrupt feature matching, dense correspondence, camera parameter estimation, and downstream 3D reconstruction tasks such as novel view synthesis. To address this problem, we propose CompMVR, a compression-aware multi-view diffusion restoration framework that jointly refines compressed views while preserving cross-view consistency, and construct compressed datasets for systematic evaluation. Unlike conventional restoration methods that optimize each image independently for 2D perceptual quality, CompMVR leverages learned compression priors and cross-view correspondence to recover visual details while improving geometric reliability across views. Experiments across diverse datasets, codecs, and compression levels demonstrate gains in restoration quality and downstream 3D tasks, including novel-view synthesis, camera pose estimation, and view matching. These results establish compressed multi-view restoration as a distinct problem and highlight its importance for 3D reconstruction under lossy compression. Code and data for this paper are available at https://mchaewon.github.io/CompMVR-project-page/.