Hi, thank you for releasing this impressive project and codebase.
I noticed that the paper mentions leveraging GPT-4V/VLM to estimate object materials from the input image, with users able to override the estimated parameters. However, I could not find the corresponding material-estimation/VLM code in the current repository. It seems that the released examples use predefined material_type and physical parameters in the YAML config files instead.
Could you kindly let me know whether the GPT-4V/VLM-based material estimation code is included somewhere in the repository, or if it is not planned to be released? Any pointers would be greatly appreciated.
Thank you very much!
Hi, thank you for releasing this impressive project and codebase.
I noticed that the paper mentions leveraging GPT-4V/VLM to estimate object materials from the input image, with users able to override the estimated parameters. However, I could not find the corresponding material-estimation/VLM code in the current repository. It seems that the released examples use predefined material_type and physical parameters in the YAML config files instead.
Could you kindly let me know whether the GPT-4V/VLM-based material estimation code is included somewhere in the repository, or if it is not planned to be released? Any pointers would be greatly appreciated.
Thank you very much!