Pixel-to-point de-projection + Grounding Dino + SAM

mbodiai / embodied-agents

Seamlessly integrate state-of-the-art transformer models into robotics stacks

https://mbodi.ai/

Apache License 2.0

112 stars 13 forks source link

Open sebbyjp opened 2 weeks ago

sebbyjp commented 2 weeks ago

Requires: Aruco marker, RGBD map, camera intrinsics matrix.

Follow-on work: RANSAC plane segmentation

Deliverable: Subclass SensoryAgent and return a Sample subclass representing the objects and poses