Color gelbooru images
An unified model, it colors the input image based on the color tags.
There are color-related matrices that can be used as conditions.
Keywords
- pixel space model
- U-shaped architecture
- CNN encoder blocks
- transformer decoder blocks
- cross-attention between encoder and decoder blocks
- 2D RoPE with center at origin
- position embedding with both cropped and non-square images
Source data
- 150k augmented, colored images
- realistic render (similar to academic art movement)
- noripro
- vspo!
References
- 2204.00631
- 2508.02324
- 2511.13720
- 2603.15031