search

sen-mao / SuppressEOT

Official Implementations "Get What You Want, Not What You Don't: Image Content Suppression for Text-to-Image Diffusion Models" (ICLR2024)
https://arxiv.org/abs/2402.05375
41 stars 1 forks source link

How can I suppress objects in generated images? #2

Open ZichengDuan opened 4 months ago

ZichengDuan commented 4 months ago

Hi Senmao, thanks for providing the code of your great work, it's really impressive and inspiring. I wonder if you can provide some instructions on how to suppress the objects or object semantics during the image generation process as in paper figure 10(b)? Thanks for your help!

sen-mao commented 4 months ago

Figure 10 does not seem to have (b).

  1. Please refer to "Suppression for generated image" in the README for suppression semantics (e.g., Monet style or ID) for generation images.

  2. Please refer to "Suppression for real image" in the README, replacing the type 'Real-Image' with 'Generated-Image' for suppressing objects in generated images