Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.19238 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918459436171264 |
|---|---|
| author | Wang, Zihan Huang, Xudong Qiao, Junbo Li, Wei Hu, Jie Chen, Xinghao Lin, Shaohui |
| author_facet | Wang, Zihan Huang, Xudong Qiao, Junbo Li, Wei Hu, Jie Chen, Xinghao Lin, Shaohui |
| contents | Real-world image super-resolution (Real-SR) has been revolutionized by leveraging the powerful generative priors of large-scale diffusion and flow-based models. However, fine-tuning these models on limited LR-HR pairs often precipitates "prior collapse" that the model sacrifices its inherent generative richness to overfit specific training degradations. This issue is further exacerbated in one-step generation, where the absence of multi-step refinement leads to significant trajectory drift and artifact generation. In this paper, we propose Allo{SR}$^2$, a novel framework that rectifies one-step SR trajectories via allomorphic generative flows to maintain high-fidelity generative realism. Specifically, we utilize Signal-to-Noise Ratio (SNR) Guided Trajectory Initialization to establish a physically grounded starting state by aligning the degradation level of LR latent features with the optimal anchoring timestep of the pre-trained flow. To ensure a stable, curvature-free path for one-step inference, we propose Flow-Anchored Trajectory Consistency (FATC), which enforces velocity-level supervision across intermediate states. Furthermore, we develop Allomorphic Trajectory Matching (ATM), a self-adversarial alignment strategy that minimizes the distributional discrepancy between the SR flow and the generative flow in a unified vector field. Extensive experiments on both synthetic and real-world benchmarks demonstrate that Allo{SR}$^2$ achieves state-of-the-art performance in one-step Real-SR, offering a superior balance between restoration fidelity and generative realism while maintaining extreme efficiency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_19238 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Allo{SR}$^2$: Rectifying One-Step Super-Resolution to Stay Real via Allomorphic Generative Flows Wang, Zihan Huang, Xudong Qiao, Junbo Li, Wei Hu, Jie Chen, Xinghao Lin, Shaohui Computer Vision and Pattern Recognition Real-world image super-resolution (Real-SR) has been revolutionized by leveraging the powerful generative priors of large-scale diffusion and flow-based models. However, fine-tuning these models on limited LR-HR pairs often precipitates "prior collapse" that the model sacrifices its inherent generative richness to overfit specific training degradations. This issue is further exacerbated in one-step generation, where the absence of multi-step refinement leads to significant trajectory drift and artifact generation. In this paper, we propose Allo{SR}$^2$, a novel framework that rectifies one-step SR trajectories via allomorphic generative flows to maintain high-fidelity generative realism. Specifically, we utilize Signal-to-Noise Ratio (SNR) Guided Trajectory Initialization to establish a physically grounded starting state by aligning the degradation level of LR latent features with the optimal anchoring timestep of the pre-trained flow. To ensure a stable, curvature-free path for one-step inference, we propose Flow-Anchored Trajectory Consistency (FATC), which enforces velocity-level supervision across intermediate states. Furthermore, we develop Allomorphic Trajectory Matching (ATM), a self-adversarial alignment strategy that minimizes the distributional discrepancy between the SR flow and the generative flow in a unified vector field. Extensive experiments on both synthetic and real-world benchmarks demonstrate that Allo{SR}$^2$ achieves state-of-the-art performance in one-step Real-SR, offering a superior balance between restoration fidelity and generative realism while maintaining extreme efficiency. |
| title | Allo{SR}$^2$: Rectifying One-Step Super-Resolution to Stay Real via Allomorphic Generative Flows |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.19238 |