search

ZiyuGuo99 / Point-Bind_Point-LLM

Align 3D Point Cloud with Multi-modalities for Large Language Models
MIT License
400 stars 31 forks source link

Questions about 3D Q&A #6

Open YueWuHKUST opened 12 months ago

YueWuHKUST commented 12 months ago

Could you provide a script or instructions about how to perform 3D Q&A given point cloud?

Thanks

pianogGG commented 11 months ago

any update about this?

ZiyuGuo99 commented 11 months ago

@YueWuHKUST @pianogGG Thanks for your interest. Our Point-LLM has been merged with LLaMA-Adapter, and we've previously provided simple scripts for 3D multi-modal inference with Point-LLM. You can find the 3D Q&A instructions here. Feel free to reach out if you have any questions or encounter any issues.

auniquesun commented 7 months ago

Do we need to install imagebind and llama-adapter related packages? They seems to be another big projects.

Can you isolate the point-llm as an independent api to facilitate 3D Q&A? Thanks.