Computer Vision Toolbox Model for Pose Mask R-CNN 6-DoF Object Pose Estimation

Pre-trained Pose Mask R-CNN model for 6-DoF object pose estimation that supports inference and transfer learning.

The Pose Mask R-CNN 6-DoF Object Pose Estimation model is based upon the existing Mask R-CNN deep learning model available in the Computer Vision Toolbox. It is trained on a dataset created using MathWorks tools. This training dataset, generated using Simulink, consists of about 5000 images and depth-maps of machine parts lying randomly in a bin, viewed from different angles and under various lighting conditions.