Computer Graphics Forum 2020

119 papers
Computer Graphics Forum published 164 papers in 2020, highlighting advancements in neural scene representations and geometry processing. Research themes focused on making neural representations more controllable, editable, and scalable for large-scale unbounded environments. Geometry-related work emphasized spectral geometry for invariant shape matching and methods for maintaining robust shape correspondence despite topological noise. Additionally, the year saw developments in the automated synthesis of 3D environments and the use of generative diffusion models for diverse human motion synthesis. Other research directions included the optimization of non-circular gears, curve skeleton extraction from point clouds, and techniques for minimizing numerical dissipation in shock-capturing simulations. Prolific authors contributing to this volume included Derek Bradley, Thabo Beeler, Emanuele Rodolà, and Maks Ovsjanikov. The collected works demonstrate a shift toward integrating deep learning with traditional graphics tasks, such as ambient occlusion generation from RGB images and differentiable forward kinematics for articulated pose regression.

Top Research Directions

Research directions most represented by Computer Graphics Forum 2020 papers.

Generative Simulation for Scaling Embodied AI 2
Spatio-Temporal Bayesian Fusion for 3D Semantic Mapping 1
Uniform Point Distribution Regularization for 3D Point Clouds 1
Non-Stationary Texture Synthesis 1
Shape Abstraction Using Parameterized Geometric Primitives 1
Iterative Normal-Vertex Refinement 1
Mitigating Numerical Dissipation in Fluid Dynamics 1
Editing Neural Implicit Representations 1
Kinematic Error Propagation in Articulated Structures 1
Morphological and Geometric Refinement of Segmentation Masks 1
Beyond Point Estimates for Parameter Uncertainty Quantification 1
From Neural Representations to Explicit Triangle Meshes 1
Differentiable Topological Losses for Deep Learning 1
Tracking Structural Features with Persistent Homology 1
Mesh Quality Sensitivity in Numerical Solvers 1

Top Authors

AuthorPapers
Derek Bradley4
Daniel A. Keim3
Derek Nowrouzezahrai3
Elmar Eisemann3
Emanuele Rodolà3
Hujun Bao3
Maks Ovsjanikov3
Tamy Boubekeur3
Zhigang Deng3
Adrien Gruson2
Andreas Kolb2
Hong Qin2
Hongyan Liu2
Jacob Munkberg2
Jon Hasselgren2
Jun He2
Justin Solomon2
Kartic Subr2
Ligang Liu2
Luca Cosmo2

Impactful Papers

Papers from Computer Graphics Forum 2020 that shaped research clusters.

Foundational

Influential

Papers

Title
Simultaneous Multi-Attribute Image-to-Image Translation Using Parallel Latent Transform Networks
Spectral Mesh Simplification
Spherical Gaussian-based Lightcuts for Glossy Interreflections
SRF-Net: Spatial Relationship Feature Network for Tooth Point Cloud Classification
SRNet: A 3D Scene Recognition Network using Static Graph and Dense Semantic Fusion
State of the Art in Time-Dependent Flow Topology: Interpreting Physical Meaningfulness Through Mathematical Properties
State of the Art on Neural Rendering
Statistics-based Motion Synthesis for Social Conversations
Subdivision-Specialized Linear Algebra Kernels for Static and Dynamic Mesh Connectivity on the GPU
Sublinear Time Force Computation for Big Complex Network Visualization
Survey on Individual Differences in Visualization
Temporal Upsampling of Point Cloud Sequences by Optimal Transport for Plant Growth Visualization
The Matchstick Model for Anisotropic Friction Cones
Topology-Aware Surface Reconstruction for Point Clouds
Two-stage Photograph Cartoonization via Line Tracing
Visual Analysis of Missing Values in Longitudinal Cohort Study Data
Visualizing Dynamics of Urban Regions Through a Geo-Semantic Graph-Based Method
v-plots: Designing Hybrid Charts for the Comparative Analysis of Data Distributions
Weakly Supervised Part-wise 3D Shape Reconstruction from Single-View RGB Images

« prev Page 3/3