Re:Real-time image editing using latent consistency models
by dvrp on Saturday November 11, 2023 @03:14AM
(#38228460)
Yup! It is different; that one is based on GAN models.
That's why you can only use what they let you paint. In our demo, you can put whatever text you want.
And we want to make it so that you can assign a (custom) label to each shape. So you can type what each shape represents.
This is the Nvidia tool: https://www.nvidia.com/en-us/studio/canvas/