EndoDepthL: Lightweight Endoscopic Monocular Depth Estimation with CNN-Transformer

TL;DR

EndoDepthL is a lightweight CNN-Transformer model designed for real-time monocular depth estimation in endoscopy, effectively handling reflections and optimizing for efficiency and accuracy.

Contribution

The paper introduces EndoDepthL, a novel lightweight architecture combining CNNs and Transformers with multi-scale and attention mechanisms for improved endoscopic depth estimation.

Findings

Achieves accurate depth estimation with a lightweight model.

Handles reflective areas effectively using confidence boundary masks.

Provides a new complexity metric for evaluating monocular depth models.

Abstract

In this study, we address the key challenges concerning the accuracy and effectiveness of depth estimation for endoscopic imaging, with a particular emphasis on real-time inference and the impact of light reflections. We propose a novel lightweight solution named EndoDepthL that integrates Convolutional Neural Networks (CNN) and Transformers to predict multi-scale depth maps. Our approach includes optimizing the network architecture, incorporating multi-scale dilated convolution, and a multi-channel attention mechanism. We also introduce a statistical confidence boundary mask to minimize the impact of reflective areas. To better evaluate the performance of monocular depth estimation in endoscopic imaging, we propose a novel complexity evaluation metric that considers network parameter size, floating-point operations, and inference frames per second. We comprehensively evaluate our proposed method and compare it with existing baseline solutions. The results demonstrate that EndoDepthL ensures depth estimation accuracy with a lightweight structure.