---
title: "SPACE: Structure-Preserving Cross-Modal Image Enhancement for Extreme Low-Light Conditions"
authors:
  - "Yue Zhang"
  - "Zhiliang Wu"
  - "Yuxuan Hou"
  - "Hehe Fan"
doi: "10.24963/ijcai.2026/235"
url: "https://ijcai.org/proceedings/2026/235"
pdf: "https://ijcai.org/proceedings/2026/0235.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "2109-2117"
track: "Main Track"
keywords:
  - "Computer Vision: Low-level Vision"
  - "Computer Vision: Multimodal learning"
  - "AI: Computer Vision"
---

# SPACE: Structure-Preserving Cross-Modal Image Enhancement for Extreme Low-Light Conditions

Yue Zhang, Zhiliang Wu, Yuxuan Hou, Hehe Fan

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 2109-2117. DOI: [10.24963/ijcai.2026/235](https://doi.org/10.24963/ijcai.2026/235)

## Abstract

Low-Light Image Enhancement (LLIE) remains challenging due to severe noise, color artifacts, and structural degradation in extreme low-light environments. Existing methods primarily rely on photometric cues and often fail to preserve geometric consistency, resulting in blurred edges and distorted textures. To address this limitation, we introduce LAMP, a large-scale dataset of Low-light Aligned Multimodal Pairs for LLIE. LAMP contains over 18k high-quality aligned pairs and consists of two complementary subsets. LAMP-Real is captured using LiDAR-based depth sensors under physically controlled illumination, while LAMP-Synthetic is generated through a physics-calibrated degradation pipeline. Both subsets provide precisely aligned RGB images, depth maps, and pixel-wise semantic annotations. Building upon LAMP, we propose SPACE, a Structure Preservation Aware Cross-modal Enhancement framework that explicitly leverages geometric priors. SPACE introduces a Depth-Adaptive HVI Transformation to decouple luminance and chrominance under depth guidance, effectively suppressing color-space noise. Furthermore, a Depth-Manifold Modulated Attention mechanism constrains feature interactions within a learned depth manifold, ensuring structural coherence during enhancement. Extensive experiments demonstrate that SPACE consistently outperforms state-of-the-art methods in visual quality and structural fidelity. The code and data are available at https://github.com/YueCheong/SPACE.