Research map / Image generation / GANs
FD-GAN: Generative Adversarial Networks with Fusion-Discriminator for Single Image Dehazing
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 adversarial loss and a standard discriminator (ablation) | Baseline FD-GAN · SOTS | PSNR: 18.5145 -> 21.6263 | Helped |
| add adversarial loss and a standard discriminator (ablation) | Baseline FD-GAN · NTIRE'18 | PSNR: 16.8574 -> 16.5912 | Hurt |
| add adversarial loss and a standard discriminator (ablation) | Baseline FD-GAN · SOTS | SSIM: 0.8714 -> 0.8922 | Helped |
| add adversarial loss and a standard discriminator (ablation) | Baseline FD-GAN · NTIRE'18 | SSIM: 0.6952 -> 0.7116 | Helped |
| add a Fusion-discriminator using both LF and HF information (ablation) | GAN dehazing model · SOTS | PSNR: 21.6263 -> 23.1529 | Helped |
| add a Fusion-discriminator using both LF and HF information (ablation) | GAN dehazing model · NTIRE'18 | PSNR: 16.5912 -> 18.0684 | Helped |
| add a Fusion-discriminator using both LF and HF information (ablation) | GAN dehazing model · SOTS | SSIM: 0.8922 -> 0.9207 | Helped |
| add a Fusion-discriminator using both LF and HF information (ablation) | GAN dehazing model · NTIRE'18 | SSIM: 0.7116 -> 0.7300 | Helped |
| use both LF and HF rather than HF alone in the discriminator (ablation) | FD-GAN · SOTS | PSNR: 21.7989 -> 23.1529 | Helped |
| use both LF and HF rather than LF alone in the discriminator (ablation) | FD-GAN · SOTS | PSNR: 22.1635 -> 23.1529 | Helped |
| train on the COCO-based synthetic dataset rather than ITS or OTS (ablation) | FD-GAN · SOTS and NTIRE'18 | PSNR and SSIM | Helped |
Closest papers by meaning
- Single Image Dehazing via Conditional Generative Adversarial Network (CVPR 2018 · similarity 0.92)
- Dehaze-RetinexGAN: Real-World Image Dehazing via Retinex-based Generative Adversarial Network (AAAI 2025 · similarity 0.91)
- Towards Domain Invariant Single Image Dehazing (AAAI 2021 · similarity 0.90)
- Learning Hazing to Dehazing: Towards Realistic Haze Generation for Real-World Image Dehazing (CVPR 2025 · similarity 0.87)
- Real-world Image Dehazing with Coherence-based Pseudo Labeling and Cooperative Unfolding Network (NeurIPS 2024 · similarity 0.87)
- PromptHaze: Prompting Real-world Dehazing via Depth Anything Model (AAAI 2025 · similarity 0.86)
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.