Computer Vision , AI

[One-page summary] Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models by Wu et al. 본문

Paper_review[short]

[One-page summary] Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models by Wu et al.

Elune001 2024. 1. 15. 21:41
Summary: ChatGPT+Prompt Managing system to use Visual Foundation Models(VFMs)
 
Approach highlight
  • Prompt managing of system principles M(P): Transform system principles into a prompt format that ChatGPT can understand.

  • Prompt managing of Foundation Model M(F): help Visual ChatGPT accurately understand and handle the task

 

  Main Results

 

   Discussion

  • Is Prompt Engineering a good answer for handling multiple tasks?