Open Gumpest opened 2 months ago
It seems that they directly used the camera segmentation renderer provided by Isaac Sim in their code. I believe they use Segment Anything in their real-world experiments.
It seems that they directly used the camera segmentation renderer provided by Isaac Sim in their code. I believe they use Segment Anything in their real-world experiments.
could you tell me the data shape for SAM?
I tried to reproduce it with SAM2, the input mask shape is uint8, binary mask, which doesn't seem to fit well with the pipeline
It is not the same with paper.