Research map / Detection and segmentation / CNN features
AOD-Net: All-In-One Dehazing Network
Boyi Li, Xiulian Peng, Zhangyang Wang, Jizheng Xu, Dan Feng
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 concatenation (multi-scale skip) layers to the K-estimation module (ablation) | AOD-Net · NYU2 TestSet A (3,170 synthetic images) | PSNR (dB): 19.0674 -> 19.6954 | Improved |
| add concatenation layers to the K-estimation module (ablation) | AOD-Net · NYU2 TestSet A (3,170 synthetic images) | SSIM: 0.7707 -> 0.8478 | Improved |
| jointly estimate 1/t(x) and A in one K(x) instead of learning t(x) with A from DCP (ablation) | AOD-Net · Challenging Real Photos (qualitative) | visual quality | Improved |
| append AOD-Net before Faster R-CNN | Faster R-CNN · Pascal VOC 2007 synthetic haze, Heavy haze | mAP: 0.5155 -> 0.5794 | Improved |
| append AOD-Net before Faster R-CNN | Faster R-CNN · Pascal VOC 2007 synthetic haze, Medium haze | mAP: 0.6046 -> 0.6401 | Improved |
| append AOD-Net before Faster R-CNN | Faster R-CNN · Pascal VOC 2007 synthetic haze, Light haze | mAP: 0.6410 -> 0.6701 | Improved |
| jointly tune AOD-Net and Faster R-CNN end to end | AOD-Net + Faster R-CNN · Pascal VOC 2007 synthetic haze, Heavy haze | mAP: 0.5794 -> 0.6819 | Improved |
| use AOD-Net instead of DehazeNet as dehazing front end | Faster R-CNN · Natural hazy images (qualitative) | detection quality | Improved |
| add heavy haze to images | Faster R-CNN · Pascal VOC 2007 synthetic haze | mAP: 0.6990 -> 0.5155 | Hurt |
| use ReLU instead of BReLU (ablation) | AOD-Net · Synthetic dehazing | performance | Improved |
| use AOD-Net instead of DCP on Middlebury | Dehazing · Middlebury TestSet B (800 synthetic images) | PSNR (dB): 18.9781 -> 21.5412 | Improved |
Later papers that compared with this method
1 later accepted paper reports AODNet as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- BidNet: Binocular Image Dehazing Without Explicit Disparity Estimation (CVPR 2020 · better on 4 of 4 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Faster R-CNN — Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (TPAMI 2016) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- FD-GAN: Generative Adversarial Networks with Fusion-Discriminator for Single Image Dehazing (AAAI 2020 · similarity 0.90)
- Proximal Dehaze-Net: A Prior Learning-Based Deep Network for Single Image Dehazing (ECCV 2018 · similarity 0.90)
- GridDehazeNet: Attention-Based Multi-Scale Network for Image Dehazing (ICCV 2019 · similarity 0.89)
- Detection-Friendly Dehazing: Object Detection in Real-World Hazy Scenes (TPAMI 2023 · similarity 0.89)
- Dehaze-RetinexGAN: Real-World Image Dehazing via Retinex-based Generative Adversarial Network (AAAI 2025 · similarity 0.87)
- Real-world Image Dehazing with Coherence-based Pseudo Labeling and Cooperative Unfolding Network (NeurIPS 2024 · similarity 0.87)
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.
