Відео

Lecture originally given as a guest lecture for the course "Exploring the Laplacian in Computer Graphics," instructed by Crane Chen at Johns Hopkins University. 00:00 Introduction 03:43 Resources 05:30 Installing Blender 06:31 Overview of Blender's interface 11:19 Navigating the viewport 13:42 Modes 16:06 Transformations 24:04 Constructing 3D shapes from basic shapes 27:34 Modelling a hammer 34...

Gravo MG is a geometric multigrid method for solving linear systems on meshes and point clouds. The technique will be presented at SIGGRAPH 2023. The Fast Forward provides a platform for paper authors to showcase their work in a 30-second video, which can be a bit light-hearted and fun. Find out more about DeltaConv on rubenwiersma.nl/gravomg

DeltaConv is a convolution layer for 3D point clouds that can be used in neural networks. The technique will be presented at SIGGRAPH 2022. The Fast Forward provides a platform for paper authors to showcase their work in a 30-second video, which can be a bit light-hearted and fun. Find out more about DeltaConv on rubenwiersma.nl/deltaconv.

Presentation for our SIGGRAPH 2022 paper on DeltaConv. Check out rubenwiersma.nl/deltaconv for more information.

This talk is concerned with a fundamental problem in geometric deep learning that arises in the construction of convolutional neural networks on surfaces. Project page: rubenwiersma.nl/hsn Paper: graphics.tudelft.nl/ruben-wiersma/papers/HSN.pdf Code: github.com/rubenwiersma/hsn

Teaser video for the presentation CNNs on Surfaces using Rotation-Equivariant Features. See the talk at SIGGRAPH 2020, from 17 August.