CoCoGaussian: Leveraging Circle of Confusion for Gaussian Splatting from Defocused Images
Journal:
arXiv
Published Date:
Dec 20, 2024
Abstract
3D Gaussian Splatting (3DGS) has attracted significant attention for its
high-quality novel view rendering, inspiring research to address real-world
challenges. While conventional methods depend on sharp images for accurate
scene reconstruction, real-world scenarios are often affected by defocus blur
due to finite depth of field, making it essential to account for realistic 3D
scene representation. In this study, we propose CoCoGaussian, a Circle of
Confusion-aware Gaussian Splatting that enables precise 3D scene representation
using only defocused images. CoCoGaussian addresses the challenge of defocus
blur by modeling the Circle of Confusion (CoC) through a physically grounded
approach based on the principles of photographic defocus. Exploiting 3D
Gaussians, we compute the CoC diameter from depth and learnable aperture
information, generating multiple Gaussians to precisely capture the CoC shape.
Furthermore, we introduce a learnable scaling factor to enhance robustness and
provide more flexibility in handling unreliable depth in scenes with reflective
or refractive surfaces. Experiments on both synthetic and real-world datasets
demonstrate that CoCoGaussian achieves state-of-the-art performance across
multiple benchmarks.
