Vanessa Wirth

I am a PhD candidate at the Chair of Visual Computing (LGDV) at FAU Erlangen-Nürnberg, supervised by Prof. Marc Stamminger.
I am part of the CRC EmpkinS.

I work at the intersection of computer vision, computer graphics, and machine learning. My research interests are focused on static and dynamic reconstruction using multiple modalities, in particular radar. Apart from theoretical research, I strive to apply my work to real-world scenarios such as medical applications.

Before my Ph.D., I studied computer science at FAU Erlangen-Nürnberg and the IDSIA lab (Prof. Dr. Jürgen Schmidhuber) at USI in Lugano (Switzerland). My master thesis was supervised by Prof. Dr. Matthias Nießner at TU Munich.

LinkedIn  /  Twitter  /  Github  /  Google Scholar  /  ResearchGate  /  Youtube

profile photo
News
  • [10/2024] 🎉 I created a Youtube channel 📹 for my oral presentations
  • [08/2024] 🎉 One paper accepted at IEEE Journal of Microwaves
  • [07/2024] 🎉 Two papers accepted (IROS, EuRAD)
  • [02/2024] 🎉 Our paper "ShaRPy" won the Digital Rheumatology Award for the most innovative publication in the field of digital rheumatology in the last two years
  • [10/2024] 🎉 Our paper "A Realistic Radar Ray Tracing Simulator for Hand Pose Imaging" won the Young Engineer Price
  • [10/2023] 🎉 Our abstract pitch about "ShaRPy" at the IEEE BHI won the Best Poster Award
  • [07/2023] 🎉 Two papers accepted (ICCVW, EuRAD)
Research

I'm interested in computer vision, computer graphics, machine learning, and novel sensor concepts. My primary application focus is within the medical domain. Representative papers are highlighted.

A Radar-Based Concept for Simultaneous High-Resolution Imaging and Pixel-Wise Velocity Analysis for Tracking Human Motion
Johanna Bräunig, Simon Heinrich, Birte Coppers, Christoph Kammel, Vanessa Wirth, Marc Stamminger, Sigrid Leyendecker, Anna-Maria Liphardt, Ingrid Ullmann, Martin Vossiek

IEEE Journal of Microwaves, 2024
[IEEE Xplore]

We propose a novel concept for high-resolution MIMO radars to track the pixel-wise velocity alongside the depth of human motions.

Automatic Spatial Calibration of Near-Field MIMO Radar With Respect to Optical Depth Sensors
Vanessa Wirth, Johanna Bräunig, Danti Khouri, Florian Gutsche, Martin Vossiek, Tim Weyrich*, Marc Stamminger*

IROS (Oral), 2024
[Project Page] [arXiv]

We propose a spatial calibration method for high-resolution radars that operate in the near-field, in conjunction with RGB-D cameras. Our calibration enables close-range sensor fusion applications with millimeter precision.

An Efficient yet High-Performance Method for Precise Radar-Based Imaging of Human Hand Poses
Johanna Bräunig, Vanessa Wirth, Marc Stamminger, Ingrid Ullmann, Martin Vossiek

EuRAD, 2024
[arXiv]

We improve our previous work for SOTA Radar reconstruction by utilizing three carrier frequencies. The proposed method enables real-time reconstructions with improved quality.

A Realistic Radar Ray Tracing Simulator for Hand Pose Imaging
Johanna Bräunig, Christian Schüssler, Vanessa Wirth, Marc Stamminger, Ingrid Ullmann, Martin Vossiek

EuRAD (Young Engineer Price), 2023
[IEEE Xplore] [arXiv]

We apply different material parameters to simulate the radar signal of a human hand using a Ray Tracing simulator. In this way, we can show that it is possible to generate very realistic simulations of radar data.

ShaRPy: Shape Reconstruction and Hand Pose Estimation from RGB-D with Uncertainty
Vanessa Wirth, Anna-Maria Liphardt, Birte Coppers, Johanna Bräunig, Simon Heinrich, Sigrid Leyendecker, Arnd Kleyer, Georg Schett, Martin Vossiek, Bernhard Egger, Marc Stamminger

ICCVW (Oral), 2023
Winner of the Digital Rheumatology Award 2024
[Project Page] [arXiv] [Code]

We propose a novel hand pose and shape tracking pipeline that calculates the uncertainty, which remains in the parameter estimates. We apply this pipeline to hand function assessments in the medical domain.

An Ulta-Efficient Approach for High-Resolution MIMO Radar Imaging of Human Hand Poses
Johanna Bräunig, Vanessa Wirth, Christoph Kammel, Christian Schüssler, Ingrid Ullmann, Marc Stamminger, Martin Vossiek

IEEE Transactions on Radar Systems, 2023
[IEEE Xplore]

We improve the performance of current SOTA Radar Imaging methods by the approximation of possible surface locations and subsequent refinement operations. Our method works with only two carrier frequencies.

Achieving Efficient and Realistic Full-Radar Simulations and Automatic Data Annotation by Exploiting Ray Meta Data from a Radar Ray Tracing Simulator
Christian Schüssler, Marcel Hoffmann, Vanessa Wirth, Björn Eskofier, Tim Weyrich, Marc Stamminger, Martin Vossiek

IEEE RadarConf, 2023
[arXiv]

We build on a realistic Range-Doppler radar signal simulator to automatically generate accurate ground-truth annotations for synthetic data.

Leisure Time Research
Author-Unification: Name-, Institution-, and Career-Sharing Co-authors
Vanessa Wirth, Vanessa Wirth

SIGBOVIK, 2023
[Link]

We investigate the phenomenon of Author-UnificAtion (AUA), which describes the high structural similarity of two co-authoring engineers that share the same forename, surname, institution, and academic career without being related by blood.


The template of this website can be stolen from Jon Barron's source code.