Published on:
25 April 2024
Primary Category:
Computer Vision and Pattern Recognition
Paper Authors:
Han Wang,
Xinning Chai,
Yiwen Wang,
Yuhong Zhang,
Rong Xie,
Li Song
Leverages powerful image generation of diffusion models
Introduces luminance guidance for color coherence
Uses high-level semantics to understand content
Designs luminance-aware decoder to enhance quality
Achieves state-of-the-art performance
Colorizing Images with Diffusion and Guidance
This paper proposes an automatic image colorization method that uses diffusion models to generate color, with guidance from the grayscale image's luminance and high-level semantics like text captions, to produce more realistic and vibrant colorization.
No comments yet, be the first to start the conversation...
Sign up to comment on this paper