stable diffusion & FPGM to OCR

隨便想到的一個概念,如果可以從雜訊到一張圖,那麼給定一張圖請他生成到最靠近的字的圖,應該就有機會可以搞成OCR問題?

Generating images with DDPMs: A PyTorch Implementation | by Brian Pulfer | MLearning.ai | Medium

--

--

aha (Cheng-Yu Lin)

Passionate about reading, gaming, and travel. Using AI to connect the physical & virtual worlds, I enjoy sharing ideas with others. Let's connect!