Skip to content

Conversation

massimilianoviola
Copy link

Hello, thanks for your fantastic open-source contributions!

While testing P3-SAM, I noticed that during demo inference in both auto_mask.py and auto_mask_no_postprocess.py, the input parameters point_num and prompt_num passed to the mesh_sam function are currently hardcoded and not affected by the provided values. This causes the inference process to run out of memory, even on high-end GPUs such as RTX 4090.

Deleting these hardcoded values allows the proper function arguments to be used, which resolves the OOM issue. Also applied the same fix to XPart for consistency.

This PR also fixes a few minor typos in the READMEs.

Thanks

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant