Department 6: Visual Computing and Artificial Intelligence

The Visual Computing and Artificial Intelligence Department investigates foundational research problems at the intersection of Computer Graphics, Computer Vision and Artificial Intelligence. It is our long term vision to develop entirely new ways to capture, represent, synthesize and simulate models of the real world at highest detail, robustness, and efficiency. To achieve this long term goal, we develop new concepts that rethink and unite established approaches from Computer Graphics and Computer Vision with concepts from Artificial Intelligence, in particular Machine Learning.

Our work lays the foundations for a new of thinking about computer graphics, and a new way of uniting and enriching the real world with computer graphics technology. It also lays the foundations for advanced methods to better perceive, understand and interpret the complex real world in motion surrounding us from visual observation, which is an essential capability of future intelligent computing systems that safely and intuitively interact with humans and the human world.

Headed by Prof. Dr. Christian Theobalt

Saarbruecken Research Center for Visual Computing, Interaction and Artificial Intelligence (result of strategic partnership between MPI-INF and Google)

Visit VIA Research Center Website

Please check out our spin-off the Captury

Visit the Captury website

Research Areas

**Prof. Dr. Christian Theobalt**
Visual Computing and Artificial Intelligence

**Dr. Mohamed Elgharib**
Vision-based Graphics

**Dr. Vladislav Golyanik**
4D and Quantum Vision

**Dr. Thomas Leimkühler**
Image Synthesis and Machine Learning

**Dr. Marc Habermann**
Graphics and Vision for Digital Humans
Head of the Real Virtual Lab

**Dr. Adam Kortylewski**
Generative Vision and Robust Learning

Spotlight and News

2024

04/2024 2 STAR reports accepted to EUROGRAPHICS 2024.
04/2024 8 papers accepted at CVPR 2024, including one poster highlight.
02/2024 4 papers accepted at 3DV 2024.

2023

10/2023 We are organizing a lecture on Advanced Topics in Neural Rendering and 3D Reconstruction in the Winter Semester 2023/2024. For more information, see here.
08/2023 Our paper "3D Gaussian Splatting for Real-time Radiance Field Rendering" received the SIGGRAPH 2023 Best Paper Award
08/2023 Our paper "HDHumans: A Hybrid Approach for High-fidelity Digital Humans" received the ACM/EUROGRAPHICS SCA 2023 Best Paper Honorable Mention.
06/2023 Marc Habermann received the Otto Hahn Medal from the Max Planck Society.
06/2023 3 papers accepted at SIGGRAPH 2023.
03/2023 14 papers accepted at CVPR 2023, including 3 Highlights.
03/2023 4 papers accepted at ICLR 2023, including 3 Spotlights.

2022

12/2022 The technology transfer organization of the Max Planck Society celebrates its anniversary in Berlin
12/2022 Our paper VoRF received the BMVC 2022 Best Paper Award Honorable Mention
09/2022 Our paper MoCapDeform received the 3DV 2022 Best Student Paper Award
09/2022 Our paper HandFlow received the VMV 2022 Best Paper Award Honorable Mention
09/2022 Marc Habermann received MVTec Dissertation Award 2022.
08/2022 11 papers accepted to ECCV 2022.
06/2022 2 papers accepted to SIGGRAPH 2022.
06/2022 MPI and Google enter into a strategic partnership to establish the Saarbruecken Center for Visual Computing, Interaction and Artificial Intelligence (link to the press).
05/2022 Ayush Tewari won Otto Hahn Medal of Max Planck Society.
05/2022 Christian Theobalt was elected fellow of the Eurographics association.
04/2022 Marc Habermann received the EUROGRAPHICS 2022 PhD Award.
04/2022 11 papers accepted to CVPR 2022.

2021

12/2021 2 papers accepted to NeurIPS 2021.
10/2021 Our work presented at ICCV 2021 was featured on the ICCV Daily.
10/2021 Franziska Müller received the Eduard Martin Award of Saarland University.
09/2021 Our paper "Learning Speech-driven 3D Conversational Gestures from Video" received the IVA 2021 Best Paper
09/2021 10 papers accepted to ICCV 2021.
07/2021 Abhimitra Meka received the EUROGRAPHICS 2021 PhD Award Honorable Mention.
06/2021 6 papers will be presented at SIGGRAPH 2021: 4 papers accepted to SIGGRAPH 2021, 2 additional ACM TOG papers will be presented.
03/2021 Christian Theobalt accepted the offer to become a Scientific Director at MPI for Informatics where he will establish a new Department working at the intersection of Graphics, Vision and AI.
03/2021 8 papers accepted to CVPR 2021, including 2 orals.
06/2021 4 papers accepted to SIGGRAPH 2021.
03/2021 Christian Theobalt accepted the offer to become a Scientific Director at MPI for Informatics where he will establish a new Department working at the intersection of Graphics, Vision and AI.
03/2021 8 papers accepted to CVPR 2021, including 2 orals.

2020

11/2020 3 paper accepted to 3D Vision 2020.
10/2020 2 paper accepted to NeurIPS 2020.
09/2020 5 papers accepted to SIGGRAPH Asia.
08/2020 Our work presented at ECCV 2020 was featured on the ECCV Daily.
08/2020 9 papers accepted at ECCV 2020, including 1 oral and 1 spotlight.
06/2020 Our paper DeepCap received the CVPR 2020 Best Student Paper Award Honorable Mention (video)
05/2020 Christian Theobalt receives EUROGRAPHICS Outstanding Technical Contributions Award.
04/2020 2 papers accepted at SIGGRAPH 2020.
04/2020 1 STAR report accepted at Eurographics 2020.
03/2020 8 papers accepted at CVPR 2020, including 3 orals.

2019

09/2019 1 paper accepted at SIGGRAPH Asia 2019.
07/2019 Mo2Cap2: Real-time Mobile 3D Motion Capture with a Cap-mounted Fisheye Camera received the Best Journal Paper Honorable Mention at IEEE TVCG Proc. VR 2019.
04/2019 5 papers presented at SIGGRAPH 2019: Three papers accepted to SIGGRAPH 2019, two additional TOG papers will be presented.
02/2019 5 papers accepted to CVPR 2019, including 3 orals.

2018

12/2018 The Chinese Olympic Team uses the marker-less motion capture system of our spin-off the Captury to prepare for the next Olympics in Tokyo. See shot put champion Chen Xiaodong being captured in action.
10/2018 FingerInput: Capturing Expressive Single-Hand Thumb-to-Finger Microgestures for On-Skin Input received the Best Academic Paper Award at the ACM International Conference on Interactive Surfaces and Spaces (ISS) 2018.
10/2018 1 paper accepted to TPAMI. High-Fidelity Monocular Face Reconstruction based on an Unsupervised Model-based Face Autoencoder
10/2018 1 paper accepted to GCPR 2018.
10/2018 1 paper accepted to SIGGRAPH Asia 2018. CurveFusion: Reconstructing Thin Structures from RGBD Sequences
10/2018 3 papers accepted to 3DV 2018.
07/2018 1 paper accepted to ECCV. A Hybrid Model for Identity Obfuscation by Face Replacement
04/2018 2 papers (+3 ACM TOG papers) accepted to SIGGRAPH. Deep Video Portraits and HeadOn: Real-time Reenactment of Human Portrait Videos
04/2018 Pablo Garrido won the Eurographics annual award for best PhD thesis (Pic).
02/2018 6 papers accepted to CVPR. More infos coming soon.
02/2018 2 STAR reports accepted to EUROGRAPHICS 2018.
02/2018 3 papers accepted to ACM TOG and will be presented at SIGGRAPH.

2017

12/2017 Our work on Real-time Hand Tracking under Occlusion from an Egocentric RGB-D Sensor won the best poster award at The 3rd Workshop on Observing and Understanding Hands in Action (HANDS 2017) .
12/2017 Christian Theobalt received the Karl Heinz Beckurts Award 2017 (Video).
11/2017 Christian Theobalt received his second ERC Grant, the ERC Consolidator Grant 4DRepLy.
11/2017 Marc Habermann won the Günter-Hotz-Medal for the best Master graduates in Computer Science.
11/2017 Dushyant Mehta and Christian Theobalt presented some of the group's works at the European Space Agency event with astronaut Matthias Maurer organized by the University of Saarland. [pic1], [pic2]
08/2017 Two papers accepted to ICCV 2017. Real-time Hand Tracking under Occlusion from an Egocentric RGB-D Sensor and MoFA:Model-based Deep Convolutional Face Autoencoder for Unsupervised Monocular Reconstruction.
07/2017 One paper accepted to ISMAR 2017. Live User-Guided Intrinsic Video For Static Scenes
07/2017 Two papers accepted to SIGGRAPH 2017. BundleFusion: Real-time Globally Consistent 3D Reconstruction using Online Surface Re-integration and VNect: Real-time 3D Human Pose Estimation with a Single RGB Camera
05/2017 Check out our PhD and Postdoc position openings here.
04/2017 Franziska Mueller is a recipient of the 2017 Google PhD Fellowship, which was covered by the local TV station.
03/2017 We are offering the Computer Vision for Computer Graphics seminar this semester.
02/2017 Franziska Mueller won one of the Frauen MINT Awards for her Master thesis on "Real-Time Hand-Object Tracking Using a Single Depth Camera". [pic1], [pic2], [pic3]
01/2017 Christian Theobalt presented the group's work to Chancellor Angela Merkel, Federal Minister Johanna Wanka, and Federel Secretary Rainer Sontowski during the the Perlen der Forschung Event in the Harnack Haus in Berlin. More information can be found on the website of the German government. [link1], [link2]. Pictures: [pic1], [pic2]

2016

12/2016 Franziska won the Günter-Hotz-Medal for the best Master graduates in Computer Science, and the "Erich-Ferdinand-Bläse-Stiftung für Forschung und Wissenschaft" award for the best Master thesis in the Faculty of Math and Computer Science. [pic1] [pic2]
11/2016 TV Report covers our work on faces.
11/2016 We are presenting our work on Real-time Dense Monocular Face Tracking of a Driver at the National IT Summit 2016 [pic1] [pic2] [pic3]
11/2016 Busy Beaver Award for Best Teaching (Seminar Computer Vision for Computer Graphics, SS2016)
10/2016 New papers at SIGGRAPH Asia 2016 and 3DV 2016.
07/2016 Best of Show Award at Emerging Technologies of SIGGRAPH 2016 for our Demo on Face2Face (link).
07/2016 Three papers accepted for ECCV2016. Details coming soon
07/2016 We demonstrated our face2face project at the popular Jimmy Kimmel Live show.
04/2016 Our work on Live Intrinsic Video will be presented at SIGGRAPH 2016.
04/2016 New paper accepted to TPAMI. Details coming soon.
04/2016 New papers at CVPR 2016, SIGGRAPH 2016, and on arXiv.
02/2016 New paper accepted to ACM Transactions on Graphics, to be presented at SIGGRAPH 2016: "Reconstruction of Personalized 3D Face Rigs from Monocular Video".

2015

03/2015 Our book Digital Representations of the Real World: How to Capture, Model, and Render Visual Reality by CRC Press that is edited by M. Magnor, O. Sorkine, O. Grau and C. Theobalt can be pre-ordered.
03/2015 Four papers accepted to CVPR 2015.

2014

12/2014 Christian will be one of the Program Chairs of 3DV in Lyon, together with Jana Kosecka and Michael Brown.
12/2014 Our paper VDub Modifying Face Video of Actors for Plausible Visual Alignment to a Dubbed Audio Track will be presented at EUROGRAPHICS 2015, project page.
12/2014 Our paper on Investigating the Dexterity of Multi-Finger Input for Mid-Air Text Entry was accepted to ACM CHI 2015, project page.
09/2014 Two papers accepted to 3DV 2014.
09/2014 Our work on Real-time Shading-based Refinement for Consumer Depth Cameras was accepted to SIGGRAPH Asia 2014, project page.
09/2014 Our work on Outdoor Human Motion Capture by Simultaneous Optimization of Pose and Camera Parameters was accepted to Computer Graphics Forum 2014, project page.
06/2014 Our work on Real-time Non-rigid Reconstruction using an RGB-D Camera was accepted to SIGGRAPH 2014, project page.
06/2014 Our work on Automatic Face Reenactment was accepted to CVPR 2014, project page.