NVlabs / FoundationPose

[CVPR 2024 Highlight] FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects
https://nvlabs.github.io/FoundationPose/
Other
1.54k stars 209 forks source link

Some questions about running run_demo. py #262

Open liguanghao1998812 opened 1 week ago

liguanghao1998812 commented 1 week ago

I have made some changes to the content of the run_demo. py program. Firstly, I connected the realsense d435i camera, and I want to use the camera to detect the pose of the target in real time. I changed the input depth image, RGB image, and mask image to automatic reading. I also aligned the depth stream and color stream, and adjusted the size. Why did the window display a bounding box that did not surround my target object after running the program? After a few seconds, the bounding box popped out of the screen, which has been bothering me for a long time. I hope everyone can help me see where the problem occurred. Thank you.

wenbowen123 commented 1 week ago

https://github.com/NVlabs/FoundationPose?tab=readme-ov-file#troubleshooting your setup might have been covered by the troubleshooting.