Task-guided Domain Gap Reduction for Monocular Depth Prediction in Endoscopy

TL;DR

This paper introduces a novel domain adaptation method for monocular depth prediction in endoscopy, effectively leveraging synthetic and real data to produce more accurate and robust depth maps for colonoscopy navigation.

Contribution

It proposes a task-guided domain gap reduction approach that selectively translates essential information between synthetic and real data, improving depth prediction accuracy.

Findings

Enhanced depth map accuracy in real colonoscopy sequences

Better generalization across different endoscopic conditions

Improved robustness over previous domain adaptation methods

Abstract

Colorectal cancer remains one of the deadliest cancers in the world. In recent years computer-aided methods have aimed to enhance cancer screening and improve the quality and availability of colonoscopies by automatizing sub-tasks. One such task is predicting depth from monocular video frames, which can assist endoscopic navigation. As ground truth depth from standard in-vivo colonoscopy remains unobtainable due to hardware constraints, two approaches have aimed to circumvent the need for real training data: supervised methods trained on labeled synthetic data and self-supervised models trained on unlabeled real data. However, self-supervised methods depend on unreliable loss functions that struggle with edges, self-occlusion, and lighting inconsistency. Methods trained on synthetic data can provide accurate depth for synthetic geometries but do not use any geometric supervisory signal from real data and overfit to synthetic anatomies and properties. This work proposes a novel approach to leverage labeled synthetic and unlabeled real data. While previous domain adaptation methods indiscriminately enforce the distributions of both input data modalities to coincide, we focus on the end task, depth prediction, and translate only essential information between the input domains. Our approach results in more resilient and accurate depth maps of real colonoscopy sequences.