Self-supervised endoscopic image key-points matching

作者：

Highlights：

• The first Self-supervised learning based approach for endoscopic image matching.

• Efficient CNN training for local endoscopic patches without need of labeled data.

• Robust data-driven descriptor against weak and highly similar textured images.

摘要

•The first Self-supervised learning based approach for endoscopic image matching.•Efficient CNN training for local endoscopic patches without need of labeled data.•Robust data-driven descriptor against weak and highly similar textured images.

论文关键词：Self-supervised learning,Feature matching,Endoscopic images,Deep learning,Image key-points matching

论文评审过程：Received 18 January 2022, Revised 11 August 2022, Accepted 24 August 2022, Available online 14 September 2022, Version of Record 26 September 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118696