mlsanityData to 14 Oct 2026

EndoMUST: Monocular Depth Estimation for Robotic Endoscopy via End-to-end Multi-step Self-supervised Training

arXiv 2506.16017 ↗First posted 19 Jun 2025cs.CV cs.ROcited by 0

Not on the map. Why?

Abstract (arXiv)

Monocular depth estimation and ego-motion estimation are significant tasks for scene perception and navigation in stable, accurate and efficient robot-assisted endoscopy. To tackle lighting variations and sparse textures in endoscopic scenes, multiple techniques including optical flow, appearance flow and intrinsic image decomposition have been introduced into the existing methods. However, the effective training strategy for multiple modules are still critical to deal with both illumination issues and information interference for self-supervised depth estimation in endoscopy. Therefore, a novel framework with multistep efficient finetuning is proposed in this work. In each epoch of end-to-end training, the process is divided into three steps, including optical flow registration, multiscale image decomposition and multiple transformation alignments. At each step, only the related networks are trained without interference of irrelevant information. Based on parameter-efficient finetuning on the foundation model, the proposed method achieves state-of-the-art performance on self-supervised depth estimation on SCARED dataset and zero-shot depth estimation on Hamlyn dataset, with 4%~10% lower error. The evaluation code of this work has been published on https://github.com/BaymaxShao/EndoMUST.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 19 of its 31 references that are in the map. First posted on arXiv 19 Jun 2025.

Builds on

19 of 31 references.

References · 19

  1. 2015Spatial Transformer Networkscited by 1,429
  2. 2024Depth Anything V2cited by 1,422
  3. 2024Depth Anything: Unleashing the Power of Large-Scale Unlabeled Datacited by 1,360
  4. 2022MonoViT: Self-Supervised Monocular Depth Estimation with a Vision Transformercited by 136
  5. 2021Stereo Correspondence and Reconstruction of Endoscopic Data Challengecited by 103
  6. 2017Occlusion Aware Unsupervised Learning of Optical Flowcited by 101
  7. 2022Lite-Mono: A Lightweight CNN and Transformer Architecture for Self-Supervised Monocular Depth Estimationcited by 100
  8. 2021Endo-Depth-and-Motion: Reconstruction and Tracking in Endoscopic Videos using Depth Networks and Photometric Constraintscited by 81
  9. 2018Structured Attention Guided Convolutional Neural Fields for Monocular Depth Estimationcited by 78
  10. 2021Self-Supervised Monocular Depth and Ego-Motion Estimation in Endoscopy: Appearance Flow to the Rescuecited by 72
  11. 2017Unsupervised Reverse Domain Adaptation for Synthetic Medical Images via Adversarial Trainingcited by 57
  12. 2019Dense Depth Estimation in Monocular Endoscopy with Self-supervised Learning Methodscited by 47
7 more
  1. 2024EndoDAC: Efficient Adapting Foundation Model for Self-Supervised Depth Estimation from Any Endoscopic Cameracited by 43
  2. 2019SLAM Endoscopy enhanced by adversarial depth predictioncited by 26
  3. 2023CGI-Stereo: Accurate and Real-Time Stereo Matching via Context and Geometry Interactioncited by 17
  4. 2018Unsupervised Odometry and Depth Learning for Endoscopic Capsule Robotscited by 14
  5. 2024MonoPCC: Photometric-invariant Cycle Constraint for Monocular Depth Estimation of Endoscopic Imagescited by 10
  6. 2018Dual CNN Models for Unsupervised Monocular Depth Estimationcited by 8
  7. 2022BDIS: Bayesian Dense Inverse Searching Method for Real-Time Stereo Surgical Image Matchingcited by 4

Built on by

Not cited yet.