DehazeNeRF: Multiple Image Haze Removal and 3D Shape Reconstruction using Neural Radiance Fields
This addresses a domain-specific problem for 3D computer vision applications in adverse weather, offering a novel solution but likely incremental as it builds on existing NeRF methods.
The paper tackles the problem of 3D vision tasks failing in hazy conditions by introducing DehazeNeRF, a framework that extends neural radiance fields with physically realistic atmospheric scattering models, enabling successful multi-view haze removal, novel view synthesis, and 3D shape reconstruction where existing methods fail.
Neural radiance fields (NeRFs) have demonstrated state-of-the-art performance for 3D computer vision tasks, including novel view synthesis and 3D shape reconstruction. However, these methods fail in adverse weather conditions. To address this challenge, we introduce DehazeNeRF as a framework that robustly operates in hazy conditions. DehazeNeRF extends the volume rendering equation by adding physically realistic terms that model atmospheric scattering. By parameterizing these terms using suitable networks that match the physical properties, we introduce effective inductive biases, which, together with the proposed regularizations, allow DehazeNeRF to demonstrate successful multi-view haze removal, novel view synthesis, and 3D shape reconstruction where existing approaches fail.