Research map / Architectures and efficiency / RNNs and normalization
PhysDNet: Physics-Driven Gradient Amplification for Real-Time Image Dehazing
Rayen Dhahri, Lei Sun, Victor J Morinigo, Alice Pagnoux, Luca Benini, Danda Pani Paudel, Luc V Gool
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Introduces PhysDNet, a lightweight multimodal dehazing network that fuses NIR imagery with sparse LiDAR depth for real-time restoration on edge devices.
Problem
Modern dehazing networks are computationally expensive and overfit to synthetic haze, limiting real-time deployment on resource-constrained edge devices.
Approach
A lightweight architecture that fuses NIR imagery with sparse LiDAR depth, with physics-driven gradient amplification; the abstract is truncated so further training details are not available.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add physics head (Koschmieder inversion) as auxiliary training head (ablation) | PhysDNet-M · STF val | PSNR: 30.02 -> 35.55 | Improved |
| add physics head (ablation) | PhysDNet-M · STF YOLOv8 detection (fog 0.06) | mAP: 10.70 -> 11.70 | Improved |
| add physics head (ablation) | PhysDNet-M · O-HAZE zero-shot | PSNR-Y: 18.92 -> 20.45 | Improved |
| add physics head (ablation) | PhysDNet-M · I-HAZE zero-shot | PSNR-Y: 16.33 -> 16.93 | Improved |
| add physics head when no transmission anchor (RGB-only, no depth) (ablation) | PhysDNet-M · SOTS-mix (RESIDE-6K) | PSNR: 28.6 -> 28.4 | No clear effect |
| use PhysDNet-L over FFA-Net+LiDAR 512 | dehazing network · STF val | PSNR: 34.79 -> 36.46 | Improved |
| add LiDAR input to DehazeFormer-T / AOD-Net at 512 | baselines · STF val | PSNR: 33.38 -> 29.91 (DehazeFormer-T 512) | Hurt |
| add physics head to NH-HAZE non-homogeneous haze (ablation) | PhysDNet-M · NH-HAZE | PSNR-Y: -0.54 dB without physics head | Improved |
Closest papers by meaning
- PHATNet: A Physics-guided Haze Transfer Network for Domain-adaptive Real-world Image Dehazing (ICCV 2025 · similarity 0.89)
- PEIE: Physics Embedded Illumination Estimation for Adaptive Dehazing (AAAI 2025 · similarity 0.87)
- Learning Hazing to Dehazing: Towards Realistic Haze Generation for Real-World Image Dehazing (CVPR 2025 · similarity 0.87)
- Advancing Real-World Image Dehazing: Perspective, Modules, and Training (TPAMI 2024 · similarity 0.87)
- Adaptive Dynamic Dehazing via Instruction-Driven and Task-Feedback Closed-Loop Optimization for Diverse Downstream Task Adaptation (AAAI 2026 · similarity 0.86)
- Exploiting Diffusion Prior for Real-World Image Dehazing with Unpaired Training (AAAI 2025 · similarity 0.85)
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.
