how to use the codebase to generate captions of a point cloud?

Pointcept / GPT4Point

[CVPR'24 Highlight] GPT4Point: A Unified Framework for Point-Language Understanding and Generation.

https://gpt4point.github.io/

MIT License

304 stars 19 forks source link

Open auniquesun opened 8 months ago

auniquesun commented 8 months ago

Thanks for sharing the paper and code. It's a great work!

My question is shown in the title, I want to generate diverse and semantic-rich captions for a specific class of point cloud, how to achieve that?

Thanks.