Revisiting the Perception-Distortion Trade-off with Spatial-Semantic Guided Super-Resolution
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917342332583936 |
|---|---|
| author | Wang, Dan Sun, Haiyan Du, Shan Wang, Z. Jane An, Zhaochong Belongie, Serge Cui, Xinrui |
| author_facet | Wang, Dan Sun, Haiyan Du, Shan Wang, Z. Jane An, Zhaochong Belongie, Serge Cui, Xinrui |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_14112 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Revisiting the Perception-Distortion Trade-off with Spatial-Semantic Guided Super-Resolution Wang, Dan Sun, Haiyan Du, Shan Wang, Z. Jane An, Zhaochong Belongie, Serge Cui, Xinrui Computer Vision and Pattern Recognition 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. |
| title | Revisiting the Perception-Distortion Trade-off with Spatial-Semantic Guided Super-Resolution |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.14112 |