Search papers, labs, and topics across Lattice.
The paper introduces the Haze-Aware Dual-Domain Refinement Network (DDR-Net) for single-image dehazing, addressing limitations in feature refinement and local structural representation found in current end-to-end networks. By employing a Haze Prior Extractor, Detail-Enhanced Blocks, and Spatial-Frequency Bottleneck Refinement, DDR-Net effectively captures multi-scale structural information and enhances feature representation. Extensive experiments show that DDR-Net surpasses existing methods on real-world benchmarks, achieving superior haze removal performance while maintaining competitive results on synthetic datasets.
DDR-Net outperforms existing dehazing methods by effectively refining features and enhancing structural representation, setting a new standard for single-image dehazing.
Single-image dehazing aims to recover clear scenes from haze-degraded images. It remains challenging due to the atmospheric scattering and the complexity of real-world haze distributions. Although recent end-to-end networks have achieved promising performance, two issues still limit their effectiveness: insufficient feature refinement at the bottleneck stage and weak local structural representation in encoder-decoder architectures. Thus, we propose a Haze-Aware Dual-Domain Refinement Network (DDR-Net) for single-image dehazing. Our method is built upon three modules: Haze Prior Extractor (HPE) provides multi-scale haze-aware priors by operating directly on downsampled hazy images; Detail-Enhanced Blocks (DE Blocks) serve as the core feature extraction units, capturing multi-scale structural information and enhancing edge and texture recovery via gradient-aware convolutions; and Spatial-Frequency Bottleneck Refinement (SFBR) at the bottleneck jointly exploits spatial and frequency information to refine bottleneck features. DDR-Net achieves more effective feature representation and reconstruction for haze removal. Extensive experiments on real-world benchmarks demonstrate that our method outperforms existing dehazing approaches. It achieves competitive performance on synthetic datasets.
