About Text conditioned Image Editing

[LeyRio]

Hello, I'm trying to use your work right now, and I want to generate an image and then edit it, like changing the color of the object in the image, etc.
The problem now is that I can generate an image with commands, but when I want to edit the objects, no image is output (as shown in figure). Is it because my prompt is not appropriate, or is there something wrong with the model, or is it because the current opensource code doesn't support image editing?
By the way, I was able to generate new images with similar content with the current model, like the demos in your paper, which means that the model should have no problems understanding and encoding the image content?

[ChocoWu]

Currently, the model does not support image editing.

[LeyRio]

Thank you for your reply, and look forward to more abundant applications of your model in the future！ At 2024-02-24 22:10:39, "Shengqiong Wu" ***@***.***> wrote: Currently, the model does not support image editing. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: ***@***.***>