Computer Graphics and Visualization › Publications

2023

Overview of our method
Lukas Uzolas, Elmar Eisemann, and Petr Kellnhofer
Template-free Articulated Neural Point Clouds for Reposable View Synthesis
In Proceedings of NeurIPS, 2023
pdf more
Simulation of balloon inflation (Model: Armadillo, 180k vertices).
Ruben Wiersma, Ahmad Nasikun, Elmar Eisemann, and Klaus Hildebrandt
A Fast Geometric Multigrid Method for Curved Surfaces
In Proceedings of SIGGRAPH (Conference Paper Track), 2023
pdf more

2022

Scenes used for our tests
Leonardo Scandolo and Elmar Eisemann
Directed acyclic graph encoding for compressed shadow maps
In Proceedings of Conference on High-Performance Graphics, 2022
pdf more
Thomas Kroes, Thomas Höllt, Elmar Eisemann, Matthias Alfeld, Francesca Gabrieli, et al.
Visual Analysis of RIS Data for Endmember Selection
In Proceedings of Eurographics Workshop on Graphics and Cultural Heritage, 2022
pdf more
Segmentation results when using DV3X, a fully connected multilayer perceptron (ANN), and the method proposed by Biermann et al. (2020) based on the floating debris index (FDI). Both the ANN and the Biermann et al. approaches achieve significantly less accurate results than DV3X.
Àlex S. Gómez, Leonardo Scandolo, and Elmar Eisemann
A learning approach for river debris detection
Int J Appl Earth Obs Geoinformation, 2022
pdf more
Mosaic image of the ground truth aligned digitisations.
Jules van der Toorn, Ruben Wiersma, Abbie Vandivere, Ricardo Marroquim, and Elmar Eisemann
A New Baseline for Feature Description on Multimodal Imaging of Paintings
In Proceedings of GCH, 2022
pdf more
(left) An image domain, (middle) Hough Transform, (right) Gaussian Sphere
Yancong Lin, Ruben Wiersma, Silvia-Laura Pintea, Klaus Hildebrandt, Elmar Eisemann, and Jan van Gemert
Deep vanishing point detection: Geometric priors make dataset variations vanish
In Proceedings of CVPR, 2022
pdf more
Ruben Wiersma, Ahmad Nasikun, Elmar Eisemann, and Klaus Hildebrandt
DeltaConv: anisotropic operators for geometric deep learning on point clouds
ACM Transactions on Graphics, 2022
pdf more
Cehao Yu, Maarten Wijntjes, Elmar Eisemann, and Sylvia Pont
Disentangling object color from illuminant color: The role of color shifts
Journal of Vision, 2022
more
Incorporating image texture information into dimensionality reduction
Alexander Vieth, Anna Vilanova, Boudewijn P. F. Lelieveldt, Elmar Eisemann, and Thomas Höllt
Incorporating Texture Information into Dimensionality Reduction for High-Dimensional Images
In Proceedings of PacificVis, 2022
pdf more
Teaser
Nejc Macek, Baran Usta, Elmar Eisemann, and Ricardo Marroquim
Real-Time Relighting of Human Faces with a Low-Cost Setup
ACM Comput Graph Interact Tech, 2022
pdf more
Yoann Coudert-Osmont, Elmar Eisemann, and Ricardo Marroquim
Semi-Automatic Perspective Lines from Paintings
In Proceedings of GCH, 2022
pdf more
Peiteng Shi, Markus Billeter, and Elmar Eisemann
Stereo-consistent screen-space ambient occlusion
ACM Comput Graph Interact Tech, 2022
pdf more
Baran Usta, Sylvia Pont, and Elmar Eisemann
Targeting Shape and Material in Lighting Design
Computer Graphics Forum, 2022
pdf more
Our method leverages a pair of spatial hierarchies over the embedding (center right) and a field (far right) over the embedding space to accelerate t-SNE minimization. Progression of minimizations (left) using these hierarchies is shown for a 60K point MNIST dataset (top) and a 1.2M point ImageNet dataset (bottom). The hierarchies are visualized for the last iteration of minimization.
Mark van de Ruit, Markus Billeter, and Elmar Eisemann
An Efficient Dual-Hierarchy t-SNE Minimization
IEEE Transactions on Visualization and Computer Graphics, 2022
pdf more

2021

A proposed interactive workflow for our mixed-initiative approach. We arbitrarily chose the steps shown in this illustration.
Mijael Bueno, Elmar Eisemann, and Rafael Bidarra
A synset-based recommender method for mixed-initiative narrative world creation
In Proceedings of ICIDS, 2021
pdf more
Left: Horizon-based ambient occlusion (HBAO) uses only a depth map and underestimates occlusion due to hidden geometry. Right: Our stochastic-depth HBAO captures occluded geometry stochastically (2ms in full HD).
Jop Vermeer, Leonardo Scandolo, and Elmar Eisemann
Stochastic-Depth Ambient Occlusion
ACM Comput Graph Interact Tech, 2021
pdf more
Our method decreases error for wavelength-dependent scattering in the presence of non-uniformly distributed emission, reflectance, and transmission throughout the scene
Mark van de Ruit and Elmar Eisemann
A multi-pass method for accelerated spectral sampling
Computer Graphics Forum, 2021
pdf more
Wineglass with dispersive dielectric materials with 32 samples per pixel. As can be seen, our reweighted schemes improve upon uniform and stratified samples and produces much smoother result in terms of color noise. Significant improvements can be observed in low discrepency sequences. All results are plotted in log-log scale.
Jerry Guo and Elmar Eisemann
Geometric Sample Reweighting for Monte Carlo Integration
Computer Graphics Forum, 2021
pdf more
Left: Visible layer transitions where bin boundaries cut through primitives. Right: Smooth layer transitions improve the results but artifacts are still visible.
Fabian Friederichs, Martin Eisemann, and Elmar Eisemann
Layered Weighted Blended Order-Independent Transparency
In Proceedings of Graphics Interface, 2021
pdf more