The repo to collect underwater image enhancement.
If you are a newcomer to the Deep Learning area and interested on underwater research, the first question you may have is "Which paper should I start reading from?"
Here is a reading roadmap of Deep Learning based underwater image enhancement and restoration!
The roadmap is constructed in accordance with the following four guidelines:
- From outline to detail
- From old to state-of-the-art
- from generic to specific areas
- focus on state-of-the-art
- Diving Deeper into Underwater Image Enhancement- A Survey
- An Underwater Image Enhancement Benchmark Dataset and Beyond; Tianjin University, Chongyi Li
- UOD: Dual Refinement Underwater Object Detection Network (Not available)
- ChinaMM (passwd: 6qci). Website; Real-world Underwater Enhancement: Challenges,Benchmarks, and Solutions
- URPC
- UDD (passwd: 2kse); Website; UDD: An Underwater Open-sea Farm Object Detection Dataset for Underwater Robot Picking; DLUT
- RUIE; Real-world Underwater Enhancement: Challenging, Benchmark and Efficient Solutions; DLUT
- EUVP; Website; Fast Underwater Image Enhancement for Improved Visual Perception
- Underwater_imagenet; Website; Enhancing Underwater Imagery using Generative Adversarial Networks(UGAN)
- UFO-120; SESR: Simultaneous Enhancement and Super-Resolution
- SUIM; Semantic Segmentation of Underwater Imagery: Dataset and Benchmark; For semantic segmentation; 7 object categories (Human divers, Aquatic plants and sea-grass, Wrecks and ruins, Robots, Reefs and invertebrates, Fish and vertebrates, Sea-floor and rocks) + Background; Train/Val (1525 pairs), Test(110 pairs)
- UIEBD; An Underwater Image Enhancement Benchmark Dataset and Beyond(Water-Net)
- U-45; A Fusion Adversarial Underwater Image Enhancement Network with a Public Test Dataset; Codes not available
- Water-GAN dataset; For WaterGAN
- SQUID dataset; Underwater Single Image Color Restoration Using Haze-Lines and a New Quantitative Dataset; Underwater-hl
- NOAA; AQUALOC: An Underwater Dataset for Visual-Inertial-Pressure Localization; Underwater, Deepsea (275m and 352m); IMU; Camera trajectory (from colmap); Magnitude
- Sea-thru; A Method For Removing Water From Underwater Images; With depth range(sparse)
- CADDY stereo data; CADDY Underwater Stereo-Vision Dataset for Human–Robot Interaction (HRI) in the Context of Diver Activities
- UWStereoNet; UWStereoNet: Unsupervised Learning for Depth Estimation and Color Correction of Underwater Stereo Imagery
- Shallow-UWnet; Shallow-UWnet : Compressed Model for Underwater Image Enhancement
- Positional Normalization; Positional Normalization; A general framework, normalization method
- All-In-One Underwater Image Enhancement using Domain-Adversarial Learning; paper; CVPRW 2019
- Underwater-hl; Diving into Haze-Lines: Color Restoration of Underwater Images; No-deep, require the depth information, matlab codes
- Funie-GAN; Fast Underwater Image Enhancement for Improved Visual Perception; real time, a good basline, proposed EUVP dataset
- Review codes; An Experimental-based Review of Image Enhancement and Image Restoration Methods for Underwater Imaging; collect some relative underwater image enhancement methods: DCP, GBdehazingRCorrection, CLAHE, ULAP, Fusion and etc.
- UWGAN_UIE; Underwater GAN for Real-world Underwater Color Restoration and Dehazing;
- UIEC^2-Net; CNN-based underwater image enhancement using two color space
- Underwater-Image-Enhancement-via-Style-Transfer; Exemplar-based Underwater Image Enhancement Augmented by Wavelet Corrected Transforms
- HybridDetectionGAN; Perceptual underwater image enhancement with deep learning and physical priors; boost the downstream detection performance, OUC-VISION group
- Deep_SESR; SESR: Simultaneous Enhancement and Super-Resolution; UFO dataset; super resolution and image enhancement
- FusionGAN (No codes)
- UEGAN; Towards Unsupervised Deep Image Enhancement with Generative Adversarial Network
- Zero-DCE; Zero-reference deep curve estimation for low-light image enhancement; CVPR 2020
- DeblurGANv2; DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Better; ICCV 2019
- CycleISP; CycleISP: Real Image Restoration via Improved Data Synthesis; Youtube; CVPR 2020 Oral
- Few-shot learning; Broader case: A Broader Study of Cross-Domain Few-Shot Learning
- Transfer learning
- Unsupervised learning
- Haifa ⭐⭐⭐⭐⭐
- UMN; ⭐⭐⭐⭐⭐
- Marine Robotics Group; MIT
- Robust Field Autonomy; Stevens Institute of Technology
- Mars
- AFRL; University of South Carolina, Columbia
- Tianjin University; UIEBD dataset
- OUC-Vision
- OUC-Ai
- Underwater Robot SLAM : Instrumentation and Frameworks
- Sonar Visual Inertial SLAM of Underwater Structures; Sonar+IMU; ICRA 2018
- SVIn2: An Underwater SLAM System using Sonar, Visual, Inertial, and Depth Sensor; Code Sonar+IMU+Depth(sparse); IROS 2019
- Contour based Reconstruction of Underwater Structures Using Sonar, Visual, Inertial, and Depth Sensor(https://afrl.cse.sc.edu/afrl/publications/public_html/papers/IROS19_1537_FI.pdf); AFRL; University of South Carolina, Columbia
- Feature-based SLAM for Imaging Sonar with Under-constrained Landmarks; ICRA 2018
- Pose-graph SLAM using Forward-looking Sonar; ICRA 2018
- Simultaneous 3D Reconstruction for Water Surface and Underwater Scene; ECCV 2018
- Fusing Concurrent Orthogonal Wide-Aperture Sonar Images for Dense Underwater 3D Reconstruction
- DeepURL: Deep Pose Estimation Framework for Underwater Relative Localization
- Dense, Sonar-based Reconstruction of Underwater Scenes
- Design and Experiments with LoCO AUV: A Low Cost Open-Source Autonomous Underwater Vehicle; UMN group
- Matching Color Aerial Images and Underwater Sonar Images using Deep Laearning for Underwater Localization
- ratslam; RatSLAM: a hippocampal model for simultaneous localization and mapping