LapSeg3D: Weakly Supervised Semantic Segmentation of Point Clouds Representing Laparoscopic Scenes

TL;DR

LapSeg3D is a semi-supervised deep learning method that efficiently segments surgical scenes in 3D point clouds, reducing manual labeling effort while maintaining high accuracy across different datasets.

Contribution

The paper introduces LapSeg3D, a novel weakly supervised approach for 3D point cloud segmentation in surgical scenes, utilizing a semi-autonomous annotation pipeline.

Findings

Achieves 0.94 F1 score for gallbladder segmentation

Generalizes well across different datasets and camera systems

Reduces manual annotation effort significantly

Abstract

The semantic segmentation of surgical scenes is a prerequisite for task automation in robot assisted interventions. We propose LapSeg3D, a novel DNN-based approach for the voxel-wise annotation of point clouds representing surgical scenes. As the manual annotation of training data is highly time consuming, we introduce a semi-autonomous clustering-based pipeline for the annotation of the gallbladder, which is used to generate segmented labels for the DNN. When evaluated against manually annotated data, LapSeg3D achieves an F1 score of 0.94 for gallbladder segmentation on various datasets of ex-vivo porcine livers. We show LapSeg3D to generalize accurately across different gallbladders and datasets recorded with different RGB-D camera systems.