Search papers, labs, and topics across Lattice.
This paper introduces Fi-Gaussian, a novel frequency-aware implicit Gaussian splatting network aimed at improving single image dehazing by effectively recovering high-frequency details and accurately modeling physical scattering. The method leverages a frequency-aware module that separates low-frequency structural information from high-frequency textures, utilizing adaptive Gaussian aggregation with complex-valued weights to enhance detail recovery. Extensive evaluations on benchmark datasets reveal that Fi-Gaussian not only achieves state-of-the-art quantitative performance but also produces visually superior dehazed images, underscoring the potential of implicit Gaussian splatting in low-level vision tasks.
Fi-Gaussian recovers fine details in dehazed images by decoupling low and high-frequency information, outperforming traditional methods that struggle with detail preservation.
Single image dehazing continues to be hindered by the loss of high-frequency details and the difficulty of accurate physical scattering modeling. To address these issues, we propose Fi-Gaussian, a frequency-aware implicit Gaussian splatting network for single image dehazing. Unlike explicit rendering methods that rely on 3D point clouds, our method employs implicit Gaussian splatting to adaptively model the underlying distribution of clear images as a continuous representation in 2D feature space. The core of the network is a frequency-aware implicit Gaussian splatting module, which decouples low-frequency structural information and high-frequency texture information in the frequency domain and then performs adaptive Gaussian aggregation with complex-valued weights to recover fine details. In addition, a physics-driven scattering renormalization mechanism is introduced to estimate the transmission map and atmospheric light under the guidance of implicit Gaussian priors. Extensive experiments on multiple benchmark datasets demonstrate that Fi-Gaussian achieves state-of-the-art quantitative performance and produces visually superior dehazed results, validating the effectiveness of implicit Gaussian splatting for low-level vision tasks.
