Revisiting the Perception-Distortion Trade-off with Spatial-Semantic Guided Super-Resolution

TL;DR

This paper introduces SpaSemSR, a diffusion-based super-resolution framework that uses spatial-semantic guidance to balance perceptual quality and fidelity, effectively reducing distortion and hallucination in high-resolution image reconstruction.

Contribution

We propose a novel spatial-semantic guided diffusion approach with dual guidance mechanisms to improve the perception-distortion trade-off in super-resolution tasks.

Findings

SpaSemSR outperforms existing methods on multiple benchmarks.

It achieves a better balance between perceptual quality and fidelity.

The framework effectively suppresses hallucinations while enhancing realism.

Abstract

Image super-resolution (SR) aims to reconstruct high resolution images with both high perceptual quality and low distortion, but is fundamentally limited by the perception-distortion trade-off. GAN-based SR methods reduce distortion but still struggle with realistic fine-grained textures, whereas diffusion-based approaches synthesize rich details but often deviate from the input, hallucinating structures and degrading fidelity. This tension raises a key challenge: how to exploit the powerful generative priors of diffusion models without sacrificing fidelity. To address this, we propose SpaSemSR, a spatial-semantic guided diffusion framework with two complementary guidances. First, spatial-grounded textual guidance integrates object-level spatial cues with semantic prompts, aligning textual and visual structures to reduce distortion. Second, semantic-enhanced visual guidance with a multi-encoder design and semantic degradation constraints unifies multimodal semantic priors, improving perceptual realism under severe degradations. These complementary guidances are adaptively fused into the diffusion process via spatial-semantic attention, suppressing distortion and hallucination while retaining the strengths of diffusion models. Extensive experiments on multiple benchmarks show that SpaSemSR achieves a superior perception-distortion balance, producing both realistic and faithful restorations.