Dual-Stage Global and Local Feature Framework for Image Dehazing

2025-08-28

AI Summary

To address the challenge of simultaneously modeling global contextual information and recovering local details in high-resolution image dehazing, this paper proposes a two-stage framework, SGLC. In the first stage, a Global Feature Generator (GFG) captures large-scale scene semantics; in the second stage, a Local Feature Enhancer (LFE) refines pixel-level texture and chromatic fidelity. SGLC introduces a model-agnostic feature fusion mechanism, enabling the first systematic global–local co-optimization—achievable via plug-and-play integration into state-of-the-art architectures such as Uformer. Evaluated on high-resolution benchmarks, SGLC achieves a significant PSNR improvement of +1.27 dB over baseline methods. It consistently restores sharp edges, natural textures, and photorealistic color reproduction, demonstrating superior visual fidelity compared to existing approaches.

Technology Category

• Computer Vision: Large Vision Models
• Search and Optimization: Local Search
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Addressing the challenge of removing atmospheric fog or haze from digital images, known as image dehazing, has recently gained significant traction in the computer vision community. Although contemporary dehazing models have demonstrated promising performance, few have thoroughly investigated high-resolution imagery. In such scenarios, practitioners often resort to downsampling the input image or processing it in smaller patches, which leads to a notable performance degradation. This drop is primarily linked to the difficulty of effectively combining global contextual information with localized, fine-grained details as the spatial resolution grows. In this chapter, we propose a novel framework, termed the Streamlined Global and Local Features Combinator (SGLC), to bridge this gap and enable robust dehazing for high-resolution inputs. Our approach is composed of two principal components: the Global Features Generator (GFG) and the Local Features Enhancer (LFE). The GFG produces an initial dehazed output by focusing on broad contextual understanding of the scene. Subsequently, the LFE refines this preliminary output by enhancing localized details and pixel-level features, thereby capturing the interplay between global appearance and local structure. To evaluate the effectiveness of SGLC, we integrated it with the Uformer architecture, a state-of-the-art dehazing model. Experimental results on high-resolution datasets reveal a considerable improvement in peak signal-to-noise ratio (PSNR) when employing SGLC, indicating its potency in addressing haze in large-scale imagery. Moreover, the SGLC design is model-agnostic, allowing any dehazing network to be augmented with the proposed global-and-local feature fusion mechanism. Through this strategy, practitioners can harness both scene-level cues and granular details, significantly improving visual fidelity in high-resolution environments.

Problem

• Addressing image dehazing for high-resolution digital imagery
• Combining global contextual information with local fine details
• Preventing performance degradation in large-scale image processing

Innovation

• Dual-stage global and local feature combination
• Model-agnostic architecture for feature fusion
• Integrated Uformer with global-local enhancement mechanism