Multi-task learning with cross-task consistency for improved depth estimation in colonoscopy

TL;DR

This paper introduces a multi-task learning framework with cross-task consistency for more accurate depth estimation in colonoscopy videos, addressing challenges posed by complex organ topology and monocular imaging.

Contribution

The work presents a novel multi-task learning model with shared encoder, attention-enhanced depth decoder, and surface normal decoder, incorporating cross-task consistency loss for improved depth estimation in colonoscopy.

Findings

14.17% reduction in relative error

10.4% improvement in $oldsymbol{ extit{ extbf{ extdelta}}}_1$ accuracy

First benchmark of state-of-the-art methods on C3VD dataset

Abstract

Colonoscopy screening is the gold standard procedure for assessing abnormalities in the colon and rectum, such as ulcers and cancerous polyps. Measuring the abnormal mucosal area and its 3D reconstruction can help quantify the surveyed area and objectively evaluate disease burden. However, due to the complex topology of these organs and variable physical conditions, for example, lighting, large homogeneous texture, and image modality estimating distance from the camera aka depth) is highly challenging. Moreover, most colonoscopic video acquisition is monocular, making the depth estimation a non-trivial problem. While methods in computer vision for depth estimation have been proposed and advanced on natural scene datasets, the efficacy of these techniques has not been widely quantified on colonoscopy datasets. As the colonic mucosa has several low-texture regions that are not well pronounced, learning representations from an auxiliary task can improve salient feature extraction, allowing estimation of accurate camera depths. In this work, we propose to develop a novel multi-task learning (MTL) approach with a shared encoder and two decoders, namely a surface normal decoder and a depth estimator decoder. Our depth estimator incorporates attention mechanisms to enhance global context awareness. We leverage the surface normal prediction to improve geometric feature extraction. Also, we apply a cross-task consistency loss among the two geometrically related tasks, surface normal and camera depth. We demonstrate an improvement of 14.17% on relative error and 10.4% improvement on $\delta_{1}$ accuracy over the most accurate baseline state-of-the-art BTS approach. All experiments are conducted on a recently released C3VD dataset; thus, we provide a first benchmark of state-of-the-art methods.