Deep Attention Based Semi-Supervised 2D-Pose Estimation for Surgical Instruments
It addresses the challenge of limited labeled data in surgical instrument pose estimation, which is incremental by combining existing semi-supervised methods with a new confidence measure.
This work tackles the problem of 2D-pose estimation for surgical instruments by applying semi-supervised learning to reduce reliance on labeled data, achieving state-of-the-art performance on benchmarks like RMIT and Endovis with a lightweight network and novel confidence measure.
For many practical problems and applications, it is not feasible to create a vast and accurately labeled dataset, which restricts the application of deep learning in many areas. Semi-supervised learning algorithms intend to improve performance by also leveraging unlabeled data. This is very valuable for 2D-pose estimation task where data labeling requires substantial time and is subject to noise. This work aims to investigate if semi-supervised learning techniques can achieve acceptable performance level that makes using these algorithms during training justifiable. To this end, a lightweight network architecture is introduced and mean teacher, virtual adversarial training and pseudo-labeling algorithms are evaluated on 2D-pose estimation for surgical instruments. For the applicability of pseudo-labelling algorithm, we propose a novel confidence measure, total variation. Experimental results show that utilization of semi-supervised learning improves the performance on unseen geometries drastically while maintaining high accuracy for seen geometries. For RMIT benchmark, our lightweight architecture outperforms state-of-the-art with supervised learning. For Endovis benchmark, pseudo-labelling algorithm improves the supervised baseline achieving the new state-of-the-art performance.