Gaussian Splatting
Gaussian splatting is a novel 3D scene representation technique that models scenes as collections of 3D Gaussian primitives, enabling real-time rendering and efficient novel view synthesis. Current research focuses on improving the accuracy and efficiency of Gaussian splatting for various applications, including virtual try-on, LiDAR simulation, and robotic perception, often incorporating techniques like multi-view consistency, depth regularization, and semantic information. This approach offers significant advantages over traditional methods in terms of speed and rendering quality, impacting fields such as autonomous driving, robotics, and virtual reality through improved scene reconstruction and manipulation.
Papers
GASP: Gaussian Splatting for Physic-Based Simulations
Piotr Borycki, Weronika Smolak, Joanna Waczyńska, Marcin Mazur, Sławomir Tadeja, Przemysław Spurek
LiDAR-3DGS: LiDAR Reinforced 3D Gaussian Splatting for Multimodal Radiance Field Rendering
Hansol Lim, Hanbeom Chang, Jongseong Brad Choi, Chul Min Yeum
Lagrangian Hashing for Compressed Neural Field Representations
Shrisudhan Govindarajan, Zeno Sambugaro, Akhmedkhan (Ahan)Shabanov, Towaki Takikawa, Daniel Rebain, Weiwei Sun, Nicola Conci, Kwang Moo Yi, Andrea Tagliasacchi
PRoGS: Progressive Rendering of Gaussian Splats
Brent Zoomers, Maarten Wijnants, Ivan Molenaers, Joni Vanherck, Jeroen Put, Lode Jorissen, Nick Michiels
GaussianPU: A Hybrid 2D-3D Upsampling Framework for Enhancing Color Point Clouds via 3D Gaussian Splatting
Zixuan Guo, Yifan Xie, Weijing Xie, Peng Huang, Fei Ma, Fei Richard Yu