Enhancing Image Super-Resolution with Optimal Boundary Conditions for Diffusion ODEs
By analyzing the optimal boundary conditions (BCs) of diffusion ODEs, we propose an approach to approximate the optimal BC and steadily generate high-quality super-resolution (SR) images from pre-trained diffusion-based SR models, outperforming existing sampling methods.
Efficient Conditional Diffusion Model with Probability Flow Sampling for Improving Image Super-Resolution
A conditional diffusion model with probability flow sampling is proposed to efficiently generate high-quality super-resolution images while maintaining consistency with low-resolution inputs.
Leveraging Diffusion Prior and Domain Shift for Efficient and High-Performance Image Super-Resolution
A novel diffusion-based super-resolution model, DoSSR, that leverages the generative power of pretrained diffusion models while significantly enhancing inference efficiency through a domain shift strategy and customized stochastic differential equation solvers.
Effective Diffusion Transformer Architecture for High-Quality Image Super-Resolution
The proposed DiT-SR architecture, which combines the advantages of U-shaped and isotropic designs, outperforms existing training-from-scratch diffusion-based super-resolution methods and even rivals the performance of prior-based methods with significantly fewer parameters.
© 2024 by Linnk AI