How to inpaint images?

[laolongboy]

Thanks. In the paper, top row of Figure 4 shows two image inpainting results. I wonder how to do that. Can I replace the image coordinates with masked images?

[idealwhite]

Wondering about the same question. It seems GPT backbone can only infer the content following forward direction, so the model could only complete half-masked images .

[LittleQBerry]

Wondering about the same question.

[cannonli7]

Thanks. In the paper, top row of Figure 4 shows two image inpainting results. I wonder how to do that. Can I replace the image coordinates with masked images?

Hello, have you figured it out how to inpaint an image with vqgan?