seg_semantic

[PopeyePxx]

Thank you for your great work! In the code you publish, you can get the "seg_semantic" directly in the camera in the simulation environment, but how do you get it in the actual deployment camera such as realsense?

[Bailey-24]

use grounded segment anything

[Everloom-129]

use grounded segment anything

could we know the data type / requirement for masks and point cloud? thanks!

[goodrain553]

it seems like the modalities of the camera are "rgb" and "depth", why is the function get_obs() able to get "segment" info?

[Sam-Firefly]

In Appendix A.5 refer that it forks this work
https://github.com/facebookresearch/segment-anything
but as this issuse say, SAM model might not be used in code.