GLOW-FM:以單步 Latent Flow Matching 高效提升低光影像
GLOW-FM: Efficient Generative Low-Light Image Enhancement with One-Step Latent Flow Matching
GLOW-FM 提出單步 latent flow-matching 框架,透過凍結的 Tiny-VAE 潛在空間進行條件傳輸,並以 Terminal Velocity Matching(TVM)只需一次函數評估,提升低光影像。實驗顯示,小型版本僅需 34 GFLOPs,影像品質具競爭力,跨數據集泛化達到最先進水平;程式碼將於論文發表後公開。
Published on Oct 22
Authors:
,
,
,
Abstract
GLOW-FM is a one-step latent flow-matching framework that enables efficient generative low-light image enhancement via single-function evaluation in a compact latent space.
Low-light image enhancement (LLIE) aims to recover well-exposed and perceptually faithful images from severely degraded observations. Recent diffusion-based generative methods achieve strong perceptual quality by leveraging powerful image priors, but their iterative sampling incurs substantial computational cost, limiting practical deployment. Despite recent efficiency gains in generative LLIE, high-quality enhancement with a single function evaluation and low overall computation remains challenging, particularly under diverse real-world low-light conditions. We introduce GLOW-FM, a one-step latent flow-matching framework for efficient generative LLIE. GLOW-FM performs conditional transport in the latent space of a frozen Tiny-VAE and leverages Terminal Velocity Matching (TVM) to enable enhancement with a single function evaluation. Matched comparisons with alternative one-step objectives support TVM as an effective design choice, and, to the best of our knowledge, this is its first adaptation to image restoration and specifically to LLIE. Extensive experiments on challenging real-world benchmarks demonstrate that GLOW-FM achieves competitive image quality and state-of-the-art cross-dataset generalization while requiring only a single function evaluation and 34 GFLOPs for the small variant. These findings highlight the potential of GLOW-FM to bring generative LLIE closer to practical deployment through single-step inference and compact latent modeling. Code will be made publicly available upon publication.
View arXiv page View PDF Project page GitHub 3 Add to collection
Get this paper in your agent:
hf papers read 2610.31222
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2610.31222 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.31222 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2610.31222 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co