CompMVR: Compression-Aware Multi-View Restoration Using Diffusion Models for Geometrically Consistent 3D Reconstruction
Paper recorded by Signals 4 on 2026-10-23 in cs.CV. Abstract reproduced from arXiv; link to the original below.
Published 2026-10-23 on arXiv · recorded by Signals 4 on 2026-10-26
Category: cs.CV · 计算机视觉 · first seen 2026-10-26
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 artifact
Read on arXiv →
Cite this page: CompMVR: Compression-Aware Multi-View Restoration Using Diffusion Models for Geometrically Consistent 3D Reconstruction: the #167 most recent of 756 cs.CV papers we have recorded (as of 2026-10-23). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/compmvr-compression-aware-multi-view-restoration-using-diffusion-models-for-geom.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable:
papers.json
Get 4 AI signals a day by email — free.
Get 4 AI signals a day by email — free