Replies: 1 comment 1 reply
-
The id you want is probably |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hi, appreciate your great work of sapien. And I hope to report a problem that seems like an issue.
I appreciate a lot if you can help!
Here is my code that I sequencly create scene and add corresponding entities inside the scene. The virtual environment is a list that contains all obj file and collision obj file. In each scene, End effector push the objects and record the final image and calculate the loss compared with GT image. In the output of each scene, the generated RGB image, point cloud, depth, video is right sequencely.
However, the mask output from camera is wrong after two sequence scene simulations. I observe the object entites ID is increasing based on the last scene, but the mask output ID is still within
0 1 2 3
.Here is my camera getting picture function. I rewrite from tutorial camera doc.
And I have also test the seg_labels[..., 0] for both mesh-level and actor level. However, the output is still wrong that mask output ID is still within
0 1 2 3
.Beta Was this translation helpful? Give feedback.
All reactions