deep_object_pose
A deep learning system for detecting known objects and estimating their 6-DoF pose from RGB images.
What is deep_object_pose?
Deep Object Pose Estimation (DOPE) is a deep learning system developed by NVIDIA for detecting known objects and estimating their 6-degree-of-freedom pose from RGB camera images. It solves the problem of enabling robots to perceive and interact with objects in 3D space without requiring depth sensors, using only visual input. The project provides a complete pipeline including data generation, training, inference, and evaluation.
Researchers and developers in robotics and computer vision who need to implement object pose estimation for applications like robotic grasping, manipulation, or augmented reality.
Developers choose DOPE for its proven performance in semantic robotic grasping, its integration with ROS for real-time applications, and its comprehensive open-source pipeline that includes training and evaluation tools alongside pre-trained models for popular datasets.
Overview
Deep Object Pose Estimation (DOPE) – ROS inference (CoRL 2018)
Use Cases
Best For
- Semantic robotic grasping of household objects
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
