3L-D: A Low-Light Level Dataset from a Sensor-Agnostic, Alignment-Preserving Acquisition Framework

Léo Rueff; Maurgène Ghezal; Valentin Gorse; Astrid Dufaure; Esteban Monsalvo; François Atteia; Ludovic Escoubas

cs.CVNov 2, 2026

3L-D: A Low-Light Level Dataset from a Sensor-Agnostic, Alignment-Preserving Acquisition Framework

Authors: Léo Rueff, Maurgène Ghezal, Valentin Gorse, Astrid Dufaure, Esteban Monsalvo, François Atteia, Ludovic Escoubas

Organizations: 3L-OPTRONICS, 2 rue Marc Donadille, 13013 Marseille, France

Abstract

Supervised low-light image and video enhancement has achieved remarkable progress, yet current models remain fundamentally constrained by the limitations of existing datasets. Prevailing methods typically rely on uncontrolled ambient captures, camera parameter manipulation (e.g., exposure or ISO adjustments), or neutral density filters, failing to replicate authentic night-sky near-infrared spectral content and genuine low-photon noise distributions. Furthermore, capturing dynamic paired sequences without introducing warping artifacts or sensor-specific biases remains an open challenge. In this paper, we introduce 3L-Dataset, a physically calibrated low-light image and video dataset generated using a modular, sensor-agnostic acquisition framework. Our test bench faithfully reproduces the standardized Midavaine night level scale from Level 1 (full moon, approximately 1000 mlx) down to Level 5 (overcast moonless night, approximately 0.1 mlx), spanning the visible to near-infrared spectrum across diverse realistic environments, including countryside, forest, urban, and desert scale models. Using high-precision programmable mechanical actuation, our acquisition framework guarantees sub-pixel spatial alignment across synchronized sextuplets, each composed of one reference image and five corresponding images acquired at the five standardized night levels, without post-processing image warping. We evaluate multiple representative deep learning baselines across varying night levels, revealing critical trade-offs between reconstruction fidelity, perceptual quality, and inference latency under severe photon starvation. The modular acquisition framework can be adapted to different cameras, optical systems, sensor technologies, and resolutions, enabling the dataset to be continuously expanded.

Figures & tables

Figure 1: Modelisation of our low-light dataset generation setup with : a) Light-proof enclosure, b) High-precision rotary stage, c) Acquisition system, d) External lightning system

Panel of Figure 2 — Figure 2: Example of an aligned sextuplet acquired from the same scene at the same temporal instant. N0 corresponds to the reference image acquired under nominal illumination, while N1 to N5 correspond to the five standardized night levels.

Figure 3: Examples of scale models created by 3L-OPTRONICS to replicate real-world environments: a) Countryside, b) Forest, c) Desert, d) City.

Figure 4: Intrinsic dimension per illumination level on raw pixels, estimated with TwoNN and MLE ( $k=5$ ), from N0 (daylight reference) to N5 (darkest). Error bars give one standard deviation over 30 bootstrap samples; asterisks mark levels where the TwoNN estimate is unstable.

Figure 5: Content preservation per illumination level on raw pixels. Left: distribution of the Pearson correlation between each frame and its pixel-aligned daylight reference (N0). Right: effective signal-to-noise ratio derived from the median correlation, $\mathrm{SNR}=\rho^{2}/(1-\rho^{2})$ ; the dashed line marks equal signal and noise power (0 dB).

Name	Light	Motion	Images/Videos	Resolution	Ground Truth	Alignment
BVI-LOWLIGHT [ 16 ]	Real	S	13,200	4240 $\times$ 2832	$\checkmark$	$\checkmark$
DUS (Static) [ 17 ]	Real	S	3,630	2160 $\times$ 1280	$\checkmark$	$\checkmark$
DUS (Dynamic) [ 17 ]	Real	D	15,000	2160 $\times$ 1280	$\times$	$\times$
DRV (Static) [ 18 ]	Real	S	20,847	5496 $\times$ 3672	$\checkmark$	$\checkmark$
DRV (Dynamic) [ 18 ]	Real	D	2,397	5496 $\times$ 3672	$\times$	$\times$
SMOID [ 5 ]	ND Filter	1D	35,800	1800 $\times$ 1000	$\checkmark$	$\checkmark$

Table 2: Comprehensive comparison of video and image datasets in low-light conditions. S and D denote static and dynamic scenes, respectively. 1D indicates a single type of motion. D* indicates stop motion .

Method	Year	Night level	PSNR (dB) $\uparrow$	SSIM $\uparrow$	LPIPS $\downarrow$	NIQE $\downarrow$	Inference time (ms) $\downarrow$
KinD [ 31 ]	2019	N3	41.98	0.9298	0.0277	16.01	87.22
		N4	39.87	0.9232	0.0584	18.47	87.22
		N5	38.41	0.9051	0.0766	19.96	87.22
Zero-DCE [ 30 ]	2020	N3	37.90	0.8693	0.7785	10.47	26.21
		N4	37.71	0.8678	0.7769	10.57	26.17
		N5	37.67	0.8671	0.7738	10.71	26.19

Table 3: Evaluation of representative low-light image enhancement networks trained on the proposed dataset at different night levels. The purpose of this benchmark is to assess the applicability of the dataset and the consistency of the results across illumination levels, rather than to rank the evaluated methods. Higher values are better for PSNR and SSIM, whereas lower values are better for LPIPS, NIQE, and inference time.

Low-Light Image Enhancement

Low-light image enhancement restores visibility, color, and detail in underexposed images while limiting noise, blur, and color distortion.

See all 98 papers

OLiVES: An Outdoor Low-Light Video Benchmark for Enhancement and Segmentation
Oct 14, 2026Yini Li, Joanne Lin, David Blackham +2Low-Light Image Enhancement Video Object Segmentation
Visual Information Laboratory University of Bristol Bristol, United Kingdom
Dark3R: Learning Structure from Motion in the Dark
Mar 5, 2026Andrew Y Guo, Anagh Malik, SaiKiran Tedla +7Structure from Motion Camera Pose Estimation
University of Toronto · Vector Institute · York University +3
ControlLight: Towards Controllable, Consistent, and Generalizable Low-Light Enhancement
May 25, 2026Yufeng Yang, Jianzhuang Liu, Jisheng Chu +4Image Enhancement Low-Light Image Enhancement
Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences · Zhejiang University

Oct 14, 2026J/K move · Enter open · S save

OLiVES: An Outdoor Low-Light Video Benchmark for Enhancement and Segmentation

Yini Li, Joanne Lin, David Blackham, David Bull+1

Visual Information Laboratory University of Bristol Bristol, United Kingdom

Low-Light Image Enhancement Video Object Segmentation

Open paper AskPDF arXiv

Low-light video processing remains an underexplored problem in computer vision, primarily due to the scarcity of high-quality ground-truth data. Existing public datasets exhibit several limitations, including restricted motion diversity, limited scene coverage, weak degradations, reliance on auxiliary modalities such as event data, and misalignment between low-light inputs and ground truth. Furthermore, current datasets typically address either low-light enhancement or scene understanding, whereas these tasks should be considered jointly. We introduce OLiVES, a benchmark dataset for outdoor low-light videos with dynamic scenes, designed to support both low-light video enhancement (LLVE) and video object segmentation (VOS). The dataset addresses key content gaps by incorporating a diverse range of natural textures, animals, occlusions, dynamic activities, and close-up scenes. To the best of our knowledge, OLiVES is the first outdoor low-light video dataset captured using professional cinema-grade cameras. It provides over 25,000 aligned normal- and low-light frame pairs, along with over 46,000 VOS annotations. We benchmark state-of-the-art LLVE and VOS methods on OLiVES and show that training on it improves performance on both tasks. The OLiVES project page is available at: https://anonymous-OLiVES.github.io/.

Abstract

Low-Light Image Enhancement

Explore similar work

OLiVES: An Outdoor Low-Light Video Benchmark for Enhancement and Segmentation

Dark3R: Learning Structure from Motion in the Dark

ControlLight: Towards Controllable, Consistent, and Generalizable Low-Light Enhancement