# Downstream Task Inspired Underwater Image Enhancement: A Perception-Aware Study from Dataset Construction to Network Design

Source: [Downstream Task Inspired Underwater Image Enhancement: A Perception-Aware Study from Dataset Construction to Network Design](<https://www.sota2.com/research/paper/69a66c662d603e6de3e3d2e2>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-04T05:00:19.274Z

Authors: Bosen Lin, Feng Gao, Yanwei Yu, Junyu Dong, Qian Du

Published: 2026-03-02T11:50:09.000Z

## Abstract

In real underwater environments, downstream image recognition tasks such as semantic segmentation and object detection often face challenges posed by problems like blurring and color inconsistencies. Underwater image enhancement (UIE) has emerged as a promising preprocessing approach, aiming to improve the recognizability of targets in underwater images. However, most existing UIE methods mainly focus on enhancing images for human visual perception, frequently failing to reconstruct high-frequency details that are critical for task-specific recognition. To address this issue, we propose a Downstream Task-Inspired Underwater Image Enhancement (DTI-UIE) framework, which leverages human visual perception model to enhance images effectively for underwater vision tasks. Specifically, we design an efficient two-branch network with task-aware attention module for feature mixing. The network benefits from a multi-stage training framework and a task-driven perceptual loss. Additionally, inspired by human perception, we automatically construct a Task-Inspired UIE Dataset (TI-UIED) using various task-specific networks. Experimental results demonstrate that DTI-UIE significantly improves task performance by generating preprocessed images that are beneficial for downstream tasks such as semantic segmentation, object detection, and instance segmentation. The codes are publicly available at https://github.com/oucailab/DTIUIE.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.01767>)

[PDF](<https://arxiv.org/pdf/2603.01767.pdf>)