GaMeS: Mesh-Based Adapting and Modification of Gaussian Splatting

1 code implementation • 2 Feb 2024 • Joanna Waczyńska, Piotr Borycki, Sławomir Tadeja, Jacek Tabor, Przemysław Spurek

In comparison, Gaussian Splatting (GS) is a novel, state-of-the-art technique for rendering points in a 3D scene by approximating their contribution to image pixels through Gaussian distributions, warranting fast training and swift, real-time rendering.

NeRFlame: FLAME-based conditioning of NeRF for 3D face rendering

1 code implementation • 10 Mar 2023 • Wojciech Zając, Joanna Waczyńska, Piotr Borycki, Jacek Tabor, Maciej Zięba, Przemysław Spurek

In contrast to traditional NeRF-based structures that use neural networks for RGB color and volume density modeling, our approach utilizes the FLAME mesh as a distinct density volume.

Hypernetwork approach to Bayesian MAML

1 code implementation • 6 Oct 2022 • Piotr Borycki, Piotr Kubacki, Marcin Przewięźlikowski, Tomasz Kuśmierczyk, Jacek Tabor, Przemysław Spurek

Unfortunately, previous modifications of MAML are limited due to the simplicity of Gaussian posteriors, MAML-like gradient-based weight updates, or by the same structure enforced for universal and adapted weights.

Few-Shot Learning