Hero-SR: One-Step Diffusion for Super-Resolution with Human Perception Priors

TL;DR

Hero-SR is a diffusion-based super-resolution framework that enhances perceptual naturalness and semantic consistency by adaptively selecting diffusion steps and integrating multi-modal guidance, achieving state-of-the-art results.

Contribution

It introduces two novel modules, DTSM and OWMS, to improve human perception alignment in super-resolution, a significant advancement over prior methods.

Findings

Achieves state-of-the-art performance in Real-SR tasks.

Effectively preserves intricate details and perceptual quality.

Demonstrates improved semantic consistency with human perception standards.

Abstract

Owing to the robust priors of diffusion models, recent approaches have shown promise in addressing real-world super-resolution (Real-SR). However, achieving semantic consistency and perceptual naturalness to meet human perception demands remains difficult, especially under conditions of heavy degradation and varied input complexities. To tackle this, we propose Hero-SR, a one-step diffusion-based SR framework explicitly designed with human perception priors. Hero-SR consists of two novel modules: the Dynamic Time-Step Module (DTSM), which adaptively selects optimal diffusion steps for flexibly meeting human perceptual standards, and the Open-World Multi-modality Supervision (OWMS), which integrates guidance from both image and text domains through CLIP to improve semantic consistency and perceptual naturalness. Through these modules, Hero-SR generates high-resolution images that not only preserve intricate details but also reflect human perceptual preferences. Extensive experiments validate that Hero-SR achieves state-of-the-art performance in Real-SR. The code will be publicly available upon paper acceptance.