Metraipolis › archive › dc_acfcc7854e2d
Flow Straight to Reality: Perceptually Consistent Flow Matching for Efficient Image Restoration — arXiv:2608.10544
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.10544
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T06:09:16.886Z
- SHA-256 of the archived text
2d7d5776e0780f8ea3206877be290a890794e16425c0d7e7c5c2e99e0c00294f- Stable URL of this page
https://metraipolis.com/documento/dc_acfcc7854e2d
How to cite
Flow Straight to Reality: Perceptually Consistent Flow Matching for Efficient Image Restoration — arXiv:2608.10544 — arXiv (en), https://arxiv.org/abs/2608.10544 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_acfcc7854e2d (Metraipolis archive dc_acfcc7854e2d, sha256 2d7d5776e0780f8ea3206877be290a890794e16425c0d7e7c5c2e99e0c00294f)
BibTeX
@misc{dcacfcc7854e2d,
title = {Flow Straight to Reality: Perceptually Consistent Flow Matching for Efficient Image Restoration — arXiv:2608.10544},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.10544},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_acfcc7854e2d; sha256 2d7d5776e0780f8ea3206877be290a890794e16425c0d7e7c5c2e99e0c00294f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10544. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Image restoration is fundamentally constrained by the tradeoff between distortion and perception: minimizing pixel-wise error yields over-smoothed results, whereas optimizing for perceptual realism often introduces structural deviations. Recent approaches attempt to balance this tradeoff via posterior sampling or multi-stage generative pipelines, yet remain computationally expensive and architecturally complex. To overcome these limitations, we propose PCFlow (Perceptually Consistent Flow Matching), a unified framework that directly parameterizes a continuous transport from degraded observations to clean targets, jointly optimizing distortion and perceptual quality. While its latent consistency flow objective drives stable and efficient few-step inference, a Latent Consistency Perceptual Loss (LCPL) imposes semantic constraints directly on the guiding velocity field, steering the dynamics toward visually sharp data manifolds. Furthermore, recognizing the inherent conflict between structural and perceptual consistencies, we integrate a conflict-free gradient projection strategy to stabilize the multi-objective optimization landscape. Combined with lightweight, convolution-only backbone, PCFlow achieves competitive performance across diverse restoration tasks at a fraction of traditional computational costs. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-12 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.10544 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-19T05:57:28.382Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).