Research map / Image generation / Image restoration
DMGINE: Day-Memory Guided Nighttime Image Enhancement for Dynamic Traffic Scenes
Ruizhou Liu, Zhe Wu, Zimo Liu, Qingfang Zheng, Qingming Huang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add foreground localization-guided contrastive learning (FLGCL) (ablation) | DMGNIE · traffic surveillance (qualitative) | foreground fidelity | Improved |
| add daytime memory (background prior) guidance (ablation) | DMGNIE · traffic surveillance (qualitative) | illumination realism | Improved |
| apply contrastive learning at deeper feature scales (S1 to S5) (ablation) | DMGNIE FLGCL · LQ-TELL | FID and LPIPS | Improved |
| apply contrastive learning only on shallow large-scale features (ablation) | DMGNIE FLGCL · LQ-TELL | FID and LPIPS | Hurt |
| use DMGNIE instead of CycleGAN | unpaired night enhancement · LQ-TELL | FID: 183.573 -> 55.505 | Improved |
| use DMGNIE instead of CycleGAN | unpaired night enhancement · LQ-TLL-QD | FID: 75.103 -> 43.181 | Improved |
| use DMGNIE instead of CUT | unpaired night enhancement · LQ-TLL-QD | NIQE: 21.826 -> 22.473 | Hurt |
| use DMGNIE instead of CUT | unpaired night enhancement · LQ-TLL-SZ | BRISQUE: 17.730 -> 20.487 | Hurt |
| finetune HVI-CIDNet on LQ-TLL | HVI-CIDNet · LQ-TELL | FID: 374.456 -> 163.282 | Improved |
| finetune Retinexformer on LQ-TLL | Retinexformer · LQ-TELL | FID: 159.928 -> 167.506 | No clear effect |
Closest papers by meaning
- Day-Night Cross-domain Vehicle Re-identification (CVPR 2024 · similarity 0.85)
- NightHaze: Nighttime Image Dehazing via Self-Prior Learning (AAAI 2025 · similarity 0.84)
- NightAdapter: Learning a Frequency Adapter for Generalizable Night-time Scene Segmentation (CVPR 2025 · similarity 0.84)
- The Parables of the Mustard Seed and the Yeast: Extremely Low-Budget, High-Performance Nighttime Semantic Segmentation (AAAI 2025 · similarity 0.84)
- NightRain: Nighttime Video Deraining via Adaptive-Rain-Removal and Adaptive-Correction (AAAI 2024 · similarity 0.84)
- DNIT: Enhancing Day-Night Image-to-Image Translation through Fine-Grained Feature Handling (Student Abstract) (AAAI 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
