NVlabs / imaginaire

NVIDIA's Deep Imagination Team's PyTorch Library
Other
4k stars 448 forks source link

Manually adding content #175

Open shehamane opened 1 year ago

shehamane commented 1 year ago

Hello! I have images with bbox coordinates, I would like the model to force the target fragments to the content distribution, not the style one. How can I do this?