Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2502.12567 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916845847576576 |
|---|---|
| author | Yang, Chao Fan, Yong Zhang, Qichao Lu, Cheng Yang, Zhijing |
| author_facet | Yang, Chao Fan, Yong Zhang, Qichao Lu, Cheng Yang, Zhijing |
| contents | Recently, the transfer application of diffusion models in super-resolu-tion tasks has faced the problem ofdecreased fidelity. Due to the inherent randomsampling characteristics ofdiffusion models, direct application in super-resolu-tion tasks can result in generated details deviating from the true distribution ofhigh-resolution images. To address this, we propose DeltaDiff, a novel frame.work that constrains the difusion process, its essence is to establish a determin-istic mapping path between HR and LR, rather than the random noise disturbanceprocess oftraditional difusion models. Theoretical analysis demonstrates a 25%reduction in diffusion entropy in the residual space compared to pixel-space diffiusion, effectively suppressing irrelevant noise interference. The experimentalresults show that our method surpasses state-of-the-art models and generates re-sults with better fidelity. This work establishes a new low-rank constrained par-adigm for applying diffusion models to image reconstruction tasks, balancingstochastic generation with structural fidelity. Our code and model are publiclyavailable at https://github.com/continueyang/DeltaDiff . |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_12567 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DeltaDiff: Reality-Driven Diffusion with AnchorResiduals for Faithful SR Yang, Chao Fan, Yong Zhang, Qichao Lu, Cheng Yang, Zhijing Computer Vision and Pattern Recognition Recently, the transfer application of diffusion models in super-resolu-tion tasks has faced the problem ofdecreased fidelity. Due to the inherent randomsampling characteristics ofdiffusion models, direct application in super-resolu-tion tasks can result in generated details deviating from the true distribution ofhigh-resolution images. To address this, we propose DeltaDiff, a novel frame.work that constrains the difusion process, its essence is to establish a determin-istic mapping path between HR and LR, rather than the random noise disturbanceprocess oftraditional difusion models. Theoretical analysis demonstrates a 25%reduction in diffusion entropy in the residual space compared to pixel-space diffiusion, effectively suppressing irrelevant noise interference. The experimentalresults show that our method surpasses state-of-the-art models and generates re-sults with better fidelity. This work establishes a new low-rank constrained par-adigm for applying diffusion models to image reconstruction tasks, balancingstochastic generation with structural fidelity. Our code and model are publiclyavailable at https://github.com/continueyang/DeltaDiff . |
| title | DeltaDiff: Reality-Driven Diffusion with AnchorResiduals for Faithful SR |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2502.12567 |