Scale-Adaptive Gaussian Splatting for Consistent Anti-Aliasing in Neural Rendering
A training-free method that applies 2D scale-adaptive filters to Gaussian primitives during inference to maintain consistent Gaussian scale across different rendering settings, enabling effective anti-aliasing through super-sampling and integration.
RoGUENeRF: A Robust Geometry-Consistent Universal Enhancer for NeRF
RoGUENeRF enhances NeRF renderings by combining 3D and 2D vision concepts, improving image quality while maintaining geometric consistency.
HourglassNeRF: A Novel Approach for Few-shot Neural Rendering
HourglassNeRF proposes a novel hourglass casting strategy for few-shot neural rendering, achieving superior results with adaptive high-frequency regularization and luminance consistency.
RoGUENeRF: Enhancing NeRF with Geometry Consistency
RoGUENeRF enhances NeRF renderings by combining 3D alignment, non-rigid refinement, and geometric attention for improved image quality.
Anisotropic Neural Representation Learning for High-Quality Neural Rendering: Improving Scene Reconstruction with Anisotropic Features
The author proposes an anisotropic neural representation using spherical harmonic functions to enhance scene reconstruction and rendering quality in NeRFs.
Global Latent Neural Rendering by Thomas Tanay and Matteo Maggioni at Huawei Noah’s Ark Lab
Global Latent Neural Rendering introduces a novel approach to view synthesis using ConvGLR, outperforming existing methods by significant margins.
© 2024 by Linnk AI