NVlabs/Deep_Object_Pose

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1178 stars · 304 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

NVIDIA's Deep Object Pose Estimation (DOPE), a deep learning system for detecting known objects and estimating their 6-DoF pose from RGB camera images. It includes training, inference, evaluation, and synthetic data generation code, plus ROS1 and ROS2 (via Isaac ROS) packages for robotics integration.

Use cases

• estimate 6-DoF pose of known objects from RGB images
• train a pose estimation model on custom objects
• generate synthetic training data for object pose estimation
• run object pose estimation in a ROS robotics pipeline
• evaluate pose estimation results on YCB or HOPE datasets
• detect and localize objects for robot manipulation

When to choose

• you need 6-DoF pose estimation of known objects from RGB cameras
• you are building a robotics application with ROS1 or ROS2
• you want to train pose estimation models on your own objects with synthetic data
• you work with YCB or HOPE benchmark datasets

When to avoid

• you need pose estimation for unknown or novel objects without training
• you require a permissively licensed model for commercial use (CC BY-NC-SA 4.0 is non-commercial)
• you need RGB-D-only or depth-based methods rather than RGB
• you lack an NVIDIA GPU for inference

Facets

library · maturity active

computer-vision machine-learning deep-learning sdk robotics computer-vision artificial-intelligence python 6dof-pose-estimation ros object-detection ycb hope-dataset nvidia corl-2018 linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="NVlabs/Deep_Object_Pose")