mlsanityData to 14 Oct 2026

Striving for Faster and Better: A One-Layer Architecture with Auto Re-parameterization for Low-Light Image Enhancement

arXiv 2502.19867 ↗First posted 27 Feb 2025cs.CVcited by 2

Not on the map. Why?

Abstract (arXiv)

Deep learning-based low-light image enhancers have made significant progress in recent years, with a trend towards achieving satisfactory visual quality while gradually reducing the number of parameters and improving computational efficiency. In this work, we aim to delving into the limits of image enhancers both from visual quality and computational efficiency, while striving for both better performance and faster processing. To be concrete, by rethinking the task demands, we build an explicit connection, i.e., visual quality and computational efficiency are corresponding to model learning and structure design, respectively. Around this connection, we enlarge parameter space by introducing the re-parameterization for ample model learning of a pre-defined minimalist network (e.g., just one layer), to avoid falling into a local solution. To strengthen the structural representation, we define a hierarchical search scheme for discovering a task-oriented re-parameterized structure, which also provides powerful support for efficiency. Ultimately, this achieves efficient low-light image enhancement using only a single convolutional layer, while maintaining excellent visual quality. Experimental results show our sensible superiority both in quality and efficiency against recently-proposed methods. Especially, our running time on various platforms (e.g., CPU, GPU, NPU, DSP) consistently moves beyond the existing fastest scheme. The source code will be released at https://github.com/vis-opt-group/AR-LLIE.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 37 of its 73 references that are in the map. First posted on arXiv 27 Feb 2025.

Builds on

37 of 73 references.

References · 37

  1. 2018Deep Retinex Decomposition for Low-Light Enhancementcited by 723
  2. 2021RepVGG: Making VGG-style ConvNets Great Againcited by 522
  3. 2019Kindling the Darkness: A Practical Low-light Image Enhancercited by 355
  4. 2020Retinex-inspired Unrolling with Cooperative Prior Architecture Search for Low-light Image Enhancementcited by 303
  5. 2023Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspectivecited by 280
  6. 2021Learning to Enhance Low-Light Image via Zero-Reference Deep Curve Estimationcited by 192
  7. 2019ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution Blockscited by 178
  8. 2022Ultra-High-Definition Low-Light Image Enhancement: A Benchmark and Transformer-Based Methodcited by 174
  9. 2018Getting to Know Low-light Images with The Exclusively Dark Datasetcited by 157
  10. 2017DSLR-Quality Photos on Mobile Devices with Deep Convolutional Networkscited by 156
  11. 2021Low-Light Image and Video Enhancement Using Deep Learning: A Surveycited by 148
  12. 2020Uncertainty-Aware Blind Image Quality Assessment in the Laboratory and Wildcited by 134
25 more
  1. 2023Multi-interactive Feature Learning and a Full-time Multi-modality Benchmark for Image Fusion and Segmentationcited by 129
  2. 2023Low-Light Image Enhancement with Wavelet-based Diffusion Modelscited by 127
  3. 2018DSFD: Dual Shot Face Detectorcited by 116
  4. 2021R2RNet: Low-light Image Enhancement via Real-low to Real-normal Networkcited by 106
  5. 2022LEDNet: Joint Low-light Enhancement and Deblurring in the Darkcited by 96
  6. 2023Implicit Neural Representation for Cooperative Low-light Image Enhancementcited by 92
  7. 2023Global Structure-Aware Diffusion Process for Low-Light Image Enhancementcited by 92
  8. 2021Diverse Branch Block: Building a Convolution as an Inception-like Unitcited by 87
  9. 2022Practical Blind Image Denoising via Swin-Conv-UNet and Data Synthesiscited by 86
  10. 2019UG2+ Track 2: A Collective Benchmark Effort for Evaluating and Advancing Image Understanding in Poor Visibility Environmentscited by 84
  11. 2021Towards Flexible Blind JPEG Artifacts Removalcited by 76
  12. 2024LightenDiffusion: Unsupervised Low-Light Image Enhancement with Latent-Retinex Diffusion Modelscited by 71
  13. 2022Multitask AET with Orthogonal Tangent Regularity for Dark Object Detectioncited by 59
  14. 2021Adaptive Unfolding Total Variation Network for Low-Light Image Enhancementcited by 55
  15. 2023A Task-guided, Implicitly-searched and Meta-initialized Deep Model for Image Fusioncited by 47
  16. 2025Infrared and Visible Image Fusion: From Data Compatibility to Task Adaptioncited by 46
  17. 2019STAR: A Structure and Texture Aware Retinex Modelcited by 41
  18. 2021HLA-Face: Joint High-Low Adaptation for Low Light Face Detectioncited by 28
  19. 2020Recurrent Exposure Generation for Low-Light Face Detectioncited by 27
  20. 2021Learning with Nested Scene Modeling and Cooperative Architecture Search for Low-Light Visioncited by 22
  21. 2024Motion-adaptive Separable Collaborative Filters for Blind Motion Deblurringcited by 20
  22. 2023LLDiffusion: Learning Degradation Representations in Diffusion Models for Low-Light Image Enhancementcited by 15
  23. 2018Learning Converged Propagations with Deep Prior Ensemble for Image Enhancementcited by 6
  24. 2023Learning a Single Convolutional Layer Model for Low Light Image Enhancementcited by 4
  25. 2023PEARL: Preprocessing Enhanced Adversarial Robust Learning of Image Deraining for Semantic Segmentationcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026SCI-Mamba: Unsupervised Learning based Low-Light Image Enhancement for Non-Cooperative Spacecraftcited by 0
  2. 2026Advantage-Guided Gate: Reshaping Open-Ended Reasoning for Vision-Based Spatial Intelligencecited by 0