The project will create a leading European-wide doctoral Collegium for research in Advanced Visual and Geometric Computing for 3D Capture, Display, and Fabrication (EVOCATION). The Collegium will train the next generation of creative, entrepreneurial and innovative experts who will be equipped with the necessary skills and competences to face current and future major challenges in scalable and high-fidelity geometry and material acquisition, extraction of structure and semantic information, processing, visualization, 3D display and 3D fabrication in professional and consumer applications. In the future, the ESRs will lead research and development of new visual and geometric computing methods in the widest variety of applications, ranging from industrial design to humanities, from medical training to urban assessment, and from creative industries to education methodologies. The EVOCATION network of public and private entities will be naturally multidisciplinary and multi-institutional and will: (a) promote, through domain-specific challenges, the culture of open science and multidisciplinary research applied to concrete problems of the real world, in strict cooperation with end users in engineering, science and humanities; (b) advance the state-of- the-art in geometry and material acquisition, geometry processing and semantic feature extraction, interactive visualization, computational fabrication, and high-bandwidth/3D display systems; (c) bridge complementary approaches for cost-effective data digitization, visualization, fabrication, and display through the integration of different methodologies in the 3D capture, processing and fabrication pipeline; (d) demonstrate the feasibility and efficiency of scalable cost-effective end-to-end techniques to virtually and physically capture and create objects with complex shape and appearance; (e) increase awareness of the benefits of advanced visual/geometric computing technology in both professional and consumer domains.

Funding

  • Horizon 2020 Marie Sklodowska-Curie Actions (MSCA) ITN 813170

Research Areas

  • In this area, we concentrate on algorithms that synthesize images to depict 3D models or scenes, often by simulating or approximating the physics of light.
  • Uses concepts from applied mathematics and computer science to design efficient algorithms for the reconstruction, analysis, manipulation, simulation and transmission of complex 3D models. Example applications are collision detection, reconstruction, compression, occlusion-aware surface handling and improved sampling conditions.
  • In this area, we focus on researching methods and algorithms that facilitate creation, representation, analysis and processing of 3D models.
  • The advent of consumer-grade 3D printing has recently sparked interest in fabrication-aware shape design and optimization. We use techniques from computer graphics and engineering to develop new computational tools to facilitate the design process of physical artifacts.

Publications

8 Publications found:
Image Bib Reference Publication Type
2024
Joao Afonso CardosoORCID iD
Approaching Under-Explored Image-Space Problems with Optimization
Supervisor: Michael WimmerORCID iD
Duration: April 2019 — 19. December 2024
[thesis]
PhD-Thesis
2022
Joao Afonso CardosoORCID iD, Bernhard KerblORCID iD, Lei Yang, Yury Uralsky, Michael WimmerORCID iD
Training and Predicting Visual Error for Real-Time Applications
Proceedings of the ACM on Computer Graphics and Interactive Techniques, 5(1):1-17, May 2022. [paper] [Paper Website]
Journal Paper with Conference Talk
Adam Celarek, Pedro Hermosilla-CasajusORCID iD, Bernhard KerblORCID iD, Timo Ropinski, Michael WimmerORCID iD
Gaussian Mixture Convolution Networks
In The Tenth International Conference on Learning Representations (ICLR 2022), pages 1-23. April 2022.
[paper] [Code on github]
Conference Paper
Simon Maximilian Fraiss
Construction and Visualization of Gaussian Mixture Models from Point Clouds for 3D Object Representation
[Master thesis] [poster]
Master Thesis
2021
Joao Afonso CardosoORCID iD, Nuno Goncalves, Michael WimmerORCID iD
Cost Volume Refinement for Depth Prediction
In Proceedings of the 25th International Conference on Pattern Recognition, pages 354-361. January 2021.
[amended-paper]
Conference Paper
2020
Dominik Hanko
Higher Hand-Drawn Detail Quality using Convolutional Assistant
Bachelor Thesis
2019
Manuel Wieser
Classification of Production Ready 2D Animation using Contour and Distance Fields
Bachelor Thesis
Error spectrum ensemble Adam Celarek, Wenzel Jakob, Michael WimmerORCID iD, Jaakko Lehtinen
Quantifying the Error of Light Transport Algorithms
Computer Graphics Forum, 38(4):111-121, July 2019. [paper_preprint] [Git repository]
Journal Paper with Conference Talk
Download list as Bibtex, HTML (Advanced, Expert), JSON (with referenced objects), CSV, Permalink

Details

Project Leader

Start Date

End Date