TL;DR
This paper introduces TADSR, a time-aware diffusion network for real-world image super-resolution that leverages timestep-conditioned generative priors for improved, controllable one-step super-resolution performance.
Contribution
The paper proposes a novel time-aware VAE encoder and VSD loss to better utilize diffusion priors at different timesteps, enabling effective, controllable one-step super-resolution.
Findings
Achieves state-of-the-art super-resolution results.
Enables controllable trade-offs between fidelity and realism.
Operates efficiently with only a single diffusion step.
Abstract
Diffusion-based real-world image super-resolution (Real-ISR) methods have demonstrated impressive performance.To achieve efficient Real-ISR, many works employ Variational Score Distillation (VSD) to distill pre-trained stable-diffusion (SD) model for one-step SR with a fixed timestep. However, since SD will perform different generative priors at different timesteps, a fixed timestep is difficult for these methods to fully leverage the generative priors in SD, leading to suboptimal performance.To address this, we propose a \textbf{T}ime-\textbf{A}ware one-step \textbf{D}iffusion Network for Real-ISR (\textbf{TADSR}). We first introduce a Time-Aware VAE Encoder, which projects the same image into different latent features based on timesteps.Through joint dynamic variation of timesteps and latent features, the student model can better align with the input pattern distribution of the…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
