Bridging Restoration and Generation Manifolds in One-Step Diffusion for Real-World Super-Resolution

TL;DR

This paper introduces IDaS-SR, a one-step diffusion-based framework for real-world image super-resolution that effectively balances restoration accuracy and perceptual quality, overcoming computational and stability challenges.

Contribution

The paper proposes MINE and CHARIOT components to address trajectory mismatches and stochastic fragility, enabling a unified, efficient super-resolution method.

Findings

IDaS-SR outperforms existing methods in real-world super-resolution tasks.

MINE accurately predicts timesteps and noise for stable inversion.

CHARIOT allows explicit control over perception and distortion trade-offs.

Abstract

Pretrained diffusion models have revolutionized real-world image super-resolution (Real-ISR) but suffer from computational bottlenecks due to iterative sampling. Recent single-step distillation accelerates inference but faces a stark perception-distortion trade-off due to rigid timestep initialization, distributional trajectory mismatches, and fragile stochastic modulation. To address this, we present Adaptive Inversion and Degradation-aware Sampling for Real-ISR (IDaS-SR), a one-step framework bridging the deterministic restoration and stochastic generation manifolds. At its core, the Manifold Inversion Noise Estimator (MINE) resolves these initialization and trajectory mismatches by predicting a severity-aware timestep and inversion noise, precisely anchoring low-quality latents onto the diffusion trajectory. Furthermore, to mitigate fragile stochastic modulation, we propose CHARIOT, a continuous generative steering mechanism. By rescheduling trajectories and interpolating noise, it enables explicit navigation of the perception-distortion boundary without compromising structural priors. Extensive experiments demonstrate that IDaS-SR outperforms state-of-the-art methods, seamlessly transitioning from a rigorous structural restorer to a sophisticated texture hallucinator in a single inference step.