4KDehazeFlow: Ultra-High-Definition Image Dehazing via Flow Matching

TL;DR

4KDehazeFlow introduces a flow matching-based approach with a haze-aware vector field and a learnable 3D LUT, achieving superior ultra-high-definition image dehazing with enhanced quality and efficiency.

Contribution

It presents a novel, architecture-agnostic dehazing method using flow matching, a learnable 3D LUT, and RK4 ODE solver for high-quality UHD image dehazing.

Findings

Outperforms seven state-of-the-art methods

Increases PSNR by 2dB on average

Better handling of dense haze and color fidelity

Abstract

Ultra-High-Definition (UHD) image dehazing faces challenges such as limited scene adaptability in prior-based methods and high computational complexity with color distortion in deep learning approaches. To address these issues, we propose 4KDehazeFlow, a novel method based on Flow Matching and the Haze-Aware vector field. This method models the dehazing process as a progressive optimization of continuous vector field flow, providing efficient data-driven adaptive nonlinear color transformation for high-quality dehazing. Specifically, our method has the following advantages: 1) 4KDehazeFlow is a general method compatible with various deep learning networks, without relying on any specific network architecture. 2) We propose a learnable 3D lookup table (LUT) that encodes haze transformation parameters into a compact 3D mapping matrix, enabling efficient inference through precomputed mappings. 3) We utilize a fourth-order Runge-Kutta (RK4) ordinary differential equation (ODE) solver to stably solve the dehazing flow field through an accurate step-by-step iterative method, effectively suppressing artifacts. Extensive experiments show that 4KDehazeFlow exceeds seven state-of-the-art methods. It delivers a 2dB PSNR increase and better performance in dense haze and color fidelity.