About this document
Real-time 3D Pose Reconstruction in Surgery by jbr5fxh9hk is a document available to read on EtoBox.
This paper presents a novel reinforcement learning approach for real-time 3D pose reconstruction of articulated surgical instruments in robotic minimally invasive surgery. The method formulates the pose estimation as a Markov Decision Process, utilizing convolutional neural networks to predict 2D joint positions and reinforcement learning to control a virtual articulated skeleton for accurate 3D alignment. Validation using semi-synthetic and real datasets demonstrates the method
- Author
- jbr5fxh9hk
- Language
- EN