Markus Worchel

@article{worchel:2023:diffparametric,
    author = {Markus Worchel and Marc Alexa},
    title = {Differentiable Rendering of Parametric Geometry},
    journal = {ACM Transactions on Graphics (Proceedings of SIGGRAPH Asia)},
    volume = {42},
    number = {6},
    year = {2023},
}

@inproceedings{worchel:2023:diff_shadow,
	title = {Differentiable Shadow Mapping for Efficient Inverse Graphics},
	author = {Markus Worchel and Marc Alexa},
	url = {https://openaccess.thecvf.com/content/CVPR2023/html/Worchel_Differentiable_Shadow_Mapping_for_Efficient_Inverse_Graphics_CVPR_2023_paper.html, CVF Open Access Version
https://mworchel.github.io/differentiable-shadow-mapping/, Project Page
https://github.com/mworchel/differentiable-shadow-mapping, Code},
	year = {2023},
	date = {2023-06-01},
	booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
	pages = {142-153},
	abstract = {We show how shadows can be efficiently generated in differentiable rendering of triangle meshes. Our central observation is that pre-filtered shadow mapping, a technique for approximating shadows based on rendering from the perspective of a light, can be combined with existing differentiable rasterizers to yield differentiable visibility information. We demonstrate at several inverse graphics problems that differentiable shadow maps are orders of magnitude faster than differentiable light transport simulation with similar accuracy -- while differentiable rasterization without shadows often fails to converge. },
	keywords = {computer graphics, differentiable rendering, machine learning, neural rendering},
	pubstate = {published},
	tppubtype = {inproceedings}
}

@inproceedings{Worchel:2022:NDS,
	title = {Multi-View Mesh Reconstruction With Neural Deferred Shading},
	author = {Markus Worchel and Rodrigo Diaz and Weiwen Hu and Oliver Schreer and Ingo Feldmann and Peter Eisert},
	url = {https://openaccess.thecvf.com/content/CVPR2022/html/Worchel_Multi-View_Mesh_Reconstruction_With_Neural_Deferred_Shading_CVPR_2022_paper.html, CVF Open Access Version
https://fraunhoferhhi.github.io/neural-deferred-shading/, Project Page
https://github.com/fraunhoferhhi/neural-deferred-shading, Code},
	year = {2022},
	date = {2022-06-01},
	booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
	pages = {6187-6197},
	abstract = {We propose an analysis-by-synthesis method for fast multi-view 3D reconstruction of opaque objects with arbitrary materials and illumination. State-of-the-art methods use both neural surface representations and neural rendering. While flexible, neural surface representations are a significant bottleneck in optimization runtime. Instead, we represent surfaces as triangle meshes and build a differentiable rendering pipeline around triangle rasterization and neural shading. The renderer is used in a gradient descent optimization where both a triangle mesh and a neural shader are jointly optimized to reproduce the multi-view images. We evaluate our method on a public 3D reconstruction dataset and show that it can match the reconstruction accuracy of traditional baselines and neural approaches while surpassing them in optimization runtime. Additionally, we investigate the shader and find that it learns an interpretable representation of appearance, enabling applications such as 3D material editing.},
	keywords = {3D reconstruction, computer graphics, computer vision, differentiable rendering, machine learning, neural rendering},
	pubstate = {published},
	tppubtype = {inproceedings}
}

@incollection{Koch:2021:HDA,
	title = {Hardware Design and Accurate Simulation of Structured-Light Scanning for Benchmarking of 3D Reconstruction Algorithms},
	author = {Sebastian Koch and Yurii Piadyk and Markus Worchel and Marc Alexa and Claudio Silva and Denis Zorin and Daniele Panozzo},
	url = {https://geometryprocessing.github.io/scanner-sim, Project Page},
	year = {2021},
	date = {2021-10-10},
	booktitle = {Thirty-fifth Conference on Neural Information Processing Systems (NeurIPS 2021)},
	issuetitle = {Datasets and Benchmarks Track},
	abstract = {Images of a real scene taken with a camera commonly differ from synthetic images of a virtual replica of the same scene, despite advances in light transport simulation and calibration. By explicitly co-developing the Structured-Light Scanning (SLS) hardware and rendering pipeline we are able to achieve negligible per-pixel difference between the real image and the synthesized image on geometrically complex calibration objects with known material properties. This approach provides an ideal test-bed for developing and evaluating data-driven algorithms in the area of 3D reconstruction, as the synthetic data is indistinguishable from real data and can be generated at large scale by simulation. We propose three benchmark challenges using a combination of acquired and synthetic data generated with our system: (1) a denoising benchmark tailored to structured-light scanning, (2) a shape completion benchmark to fill in missing data, and (3) a benchmark for surface reconstruction from dense point clouds. Besides, we provide a large collection of high-resolution scans that allow to use our system and benchmarks without reproduction of the hardware setup on our website},
	howpublished = {https://openreview.net/forum?id=bNL5VlTfe3p},
	keywords = {computer graphics},
	pubstate = {published},
	tppubtype = {incollection}
}