English

GeoDiff: Geometry-Guided Diffusion for Metric Depth Estimation

Computer Vision and Pattern Recognition 2025-10-22 v1

Abstract

We introduce a novel framework for metric depth estimation that enhances pretrained diffusion-based monocular depth estimation (DB-MDE) models with stereo vision guidance. While existing DB-MDE methods excel at predicting relative depth, estimating absolute metric depth remains challenging due to scale ambiguities in single-image scenarios. To address this, we reframe depth estimation as an inverse problem, leveraging pretrained latent diffusion models (LDMs) conditioned on RGB images, combined with stereo-based geometric constraints, to learn scale and shift for accurate depth recovery. Our training-free solution seamlessly integrates into existing DB-MDE frameworks and generalizes across indoor, outdoor, and complex environments. Extensive experiments demonstrate that our approach matches or surpasses state-of-the-art methods, particularly in challenging scenarios involving translucent and specular surfaces, all without requiring retraining.

Keywords

Cite

@article{arxiv.2510.18291,
  title  = {GeoDiff: Geometry-Guided Diffusion for Metric Depth Estimation},
  author = {Tuan Pham and Thanh-Tung Le and Xiaohui Xie and Stephan Mandt},
  journal= {arXiv preprint arXiv:2510.18291},
  year   = {2025}
}

Comments

Accepted to ICCV Findings 2025. The first two authors contributed equally. The last two authors share co-corresponding authorship

R2 v1 2026-07-01T06:57:08.170Z