Weakly Supervised Temporal Convolutional Networks for Fine-grained Surgical Activity Recognition

TL;DR

This paper presents a weakly supervised learning approach using temporal convolutional networks to recognize fine-grained surgical steps from coarser phase labels, reducing annotation effort while maintaining high accuracy.

Contribution

It introduces a step-phase dependency loss and employs a single-stage TCN with ResNet-50 for end-to-end surgical activity recognition from weakly labeled videos.

Findings

Effective on large surgical datasets

Reduces annotation effort significantly

Achieves competitive recognition accuracy

Abstract

Automatic recognition of fine-grained surgical activities, called steps, is a challenging but crucial task for intelligent intra-operative computer assistance. The development of current vision-based activity recognition methods relies heavily on a high volume of manually annotated data. This data is difficult and time-consuming to generate and requires domain-specific knowledge. In this work, we propose to use coarser and easier-to-annotate activity labels, namely phases, as weak supervision to learn step recognition with fewer step annotated videos. We introduce a step-phase dependency loss to exploit the weak supervision signal. We then employ a Single-Stage Temporal Convolutional Network (SS-TCN) with a ResNet-50 backbone, trained in an end-to-end fashion from weakly annotated videos, for temporal activity segmentation and recognition. We extensively evaluate and show the effectiveness of the proposed method on a large video dataset consisting of 40 laparoscopic gastric bypass procedures and the public benchmark CATARACTS containing 50 cataract surgeries.