MBAPose: Mask and Bounding-Box Aware Pose Estimation of Surgical Instruments with Photorealistic Domain Randomization
arXiv:2103.08105 · doi:10.1109/IROS51168.2021.9636404
Abstract
Surgical robots are usually controlled using a priori models based on the robots' geometric parameters, which are calibrated before the surgical procedure. One of the challenges in using robots in real surgical settings is that those parameters can change over time, consequently deteriorating control accuracy. In this context, our group has been investigating online calibration strategies without added sensors. In one step toward that goal, we have developed an algorithm to estimate the pose of the instruments' shafts in endoscopic images. In this study, we build upon that earlier work and propose a new framework to more precisely estimate the pose of a rigid surgical instrument. Our strategy is based on a novel pose estimation model called MBAPose and the use of synthetic training data. Our experiments demonstrated an improvement of 21 % for translation error and 26 % for orientation error on synthetic test data with respect to our previous work. Results with real test data provide a baseline for further research.
arXiv commentAccepted on IROS 2021, 8 pages
References in corpus (11)
- Generative Adversarial Networks
- DSSD : Deconvolutional Single Shot Detector
- A disciplined approach to neural network hyper-parameters: Part 1 -- learning rate, batch size, momentum, and weight decay
- Virtual Worlds as Proxy for Multi-Object Tracking Analysis
- Deep Drone Racing: From Simulation to Reality with Domain Randomization
- Synscapes: A Photorealistic Synthetic Dataset for Street Scene Parsing
- Dynamic Active Constraints for Surgical Robots using Vector Field Inequalities
- EfficientPose: An efficient, accurate and scalable end-to-end 6D multi object pose estimation approach
- A Unified Framework for the Teleoperation of Surgical Robots in Constrained Workspaces
- Single-Shot Pose Estimation of Surgical Robot Instruments' Shafts from Monocular Endoscopic Images
- Adapting Object Detectors with Conditional Domain Normalization