Augmented Radiance Field: A General Framework for Enhanced Gaussian Splatting
This work addresses a domain-specific problem in radiance field reconstruction for computer vision and graphics, offering incremental improvements to existing 3D Gaussian Splatting methods.
The paper tackles the limitation of 3D Gaussian Splatting in representing complex reflections by proposing an enhanced Gaussian kernel with view-dependent opacity and an error-driven compensation strategy, resulting in improved rendering performance and greater parameter efficiency compared to state-of-the-art NeRF methods.
Due to the real-time rendering performance, 3D Gaussian Splatting (3DGS) has emerged as the leading method for radiance field reconstruction. However, its reliance on spherical harmonics for color encoding inherently limits its ability to separate diffuse and specular components, making it challenging to accurately represent complex reflections. To address this, we propose a novel enhanced Gaussian kernel that explicitly models specular effects through view-dependent opacity. Meanwhile, we introduce an error-driven compensation strategy to improve rendering quality in existing 3DGS scenes. Our method begins with 2D Gaussian initialization and then adaptively inserts and optimizes enhanced Gaussian kernels, ultimately producing an augmented radiance field. Experiments demonstrate that our method not only surpasses state-of-the-art NeRF methods in rendering performance but also achieves greater parameter efficiency. Project page at: https://xiaoxinyyx.github.io/augs.