ColonAdapter: Geometry Estimation Through Foundation Model Adaptation for Colonoscopy

TL;DR

This paper introduces ColonAdapter, a self-supervised framework that adapts foundation models for accurate 3D geometry estimation in colonoscopy images, overcoming challenges posed by clinical scene textures and lighting.

Contribution

We propose a novel fine-tuning method with modules and loss functions tailored for colonoscopy data, improving geometric estimation accuracy without ground-truth parameters.

Findings

Achieves state-of-the-art results in camera pose estimation

Improves monocular depth prediction accuracy

Enhances dense 3D point map reconstruction

Abstract

Estimating 3D geometry from monocular colonoscopy images is challenging due to non-Lambertian surfaces, moving light sources, and large textureless regions. While recent 3D geometric foundation models eliminate the need for multi-stage pipelines, their performance deteriorates in clinical scenes. These models are primarily trained on natural scene datasets and struggle with specularity and homogeneous textures typical in colonoscopy, leading to inaccurate geometry estimation. In this paper, we present ColonAdapter, a self-supervised fine-tuning framework that adapts geometric foundation models for colonoscopy geometry estimation. Our method leverages pretrained geometric priors while tailoring them to clinical data. To improve performance in low-texture regions and ensure scale consistency, we introduce a Detail Restoration Module (DRM) and a geometry consistency loss. Furthermore, a confidence-weighted photometric loss enhances training stability in clinical environments. Experiments on both synthetic and real datasets demonstrate that our approach achieves state-of-the-art performance in camera pose estimation, monocular depth prediction, and dense 3D point map reconstruction, without requiring ground-truth intrinsic parameters.