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

2026-03-02

AI Summary

This work addresses the limitation of existing underwater image enhancement methods, which primarily optimize for human visual perception and often fail to recover high-frequency details critical for downstream vision tasks such as semantic segmentation and object detection. To bridge this gap, the authors propose DTI-UIE, a task-driven underwater image enhancement framework that jointly optimizes perceptual quality and task performance through a dual-branch architecture and a task-aware attention mechanism. The study introduces TI-UIED, the first task-oriented underwater image enhancement dataset, and designs a task-aware loss function alongside a multi-stage training strategy. Extensive experiments demonstrate that the proposed method significantly outperforms state-of-the-art approaches across multiple downstream tasks, effectively enhancing the robustness and accuracy of machine vision systems in underwater environments.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

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.

Problem

• underwater image enhancement
• downstream tasks
• perception-aware
• high-frequency details
• task-specific recognition

Innovation

• Task-Inspired Enhancement
• Underwater Image Enhancement
• Perception-Aware Network
• Task-Driven Loss
• Automated Dataset Construction