UIE
21 papers with code • 0 benchmarks • 0 datasets
Underwater image enhancement is a technique used to improve the quality of underwater images. Due to the unique properties of the underwater environment, images captured underwater often suffer from degradation caused by absorption and scattering of light. These effects can result in low contrast, blurred images with a dominant blue or green color cast. Enhancement techniques aim to correct these issues and improve the visibility within the image. These methods can include color correction to remove the color cast, contrast enhancement to improve the visibility of underwater objects, and dehazing techniques to reduce the scattering effect. These enhancements are crucial in various applications, including underwater exploration, marine biology research, underwater archaeology, and in improving the performance of underwater vision systems used in autonomous underwater vehicles.
Benchmarks
These leaderboards are used to track progress in UIE
Most implemented papers
U-shape Transformer for Underwater Image Enhancement
LintaoPeng/U-shape_Transformer
•
•
In this work, we constructed a large-scale underwater image (LSUI) dataset including 5004 image pairs, and reported an U-shape Transformer network where the transformer model is for the first time introduced to the UIE task.
Twice Mixing: A Rank Learning based Quality Assessment Approach for Underwater Image Enhancement
Our approach, termed Twice Mixing, is motivated by the observation that a mid-quality image can be generated by mixing a high-quality image with its low-quality version.
Data-driven input reconstruction and experimental validation
This paper addresses a data-driven input reconstruction problem based on Willems' Fundamental Lemma in which unknown input estimators (UIEs) are constructed directly from historical I/O data.
Unified Structure Generation for Universal Information Extraction
Information extraction suffers from its varying targets, heterogeneous structures, and demand-specific schemas.
Reinforced Swin-Convs Transformer for Underwater Image Enhancement
To address problems, a novel U-Net based Reinforced Swin-Convs Transformer for the Underwater Image Enhancement method (URSCT-UIE) is proposed.
Uncertainty Inspired Underwater Image Enhancement
After that, we adopt a consensus process to predict a deterministic result based on a set of samples from the distribution.
Underwater Ranker: Learn Which Is Better and How to Be Better
To achieve that, we also contribute a dataset, URankerSet, containing sufficient results enhanced by different UIE algorithms and the corresponding perceptual rankings, to train our URanker.
SyreaNet: A Physically Guided Underwater Image Enhancement Framework Integrating Synthetic and Real Images
Although learning-based UIE methods have made remarkable achievements in recent years, it's still challenging for them to consistently deal with various underwater conditions, which could be caused by: 1) the use of the simplified atmospheric image formation model in UIE may result in severe errors; 2) the network trained solely with synthetic images might have difficulty in generalizing well to real underwater images.
LasUIE: Unifying Information Extraction with Latent Adaptive Structure-aware Generative Language Model
chocowu/lasuie
•
•
Universally modeling all typical information extraction tasks (UIE) with one generative language model (GLM) has revealed great potential by the latest study, where various IE predictions are unified into a linearized hierarchical expression under a GLM.
CodeIE: Large Code Generation Models are Better Few-Shot Information Extractors
dasepli/codeie
•
•
A common practice is to recast the task into a text-to-text format such that generative LLMs of natural language (NL-LLMs) like GPT-3 can be prompted to solve it.
