Working in the field of computer vision at the University of Stuttgart, IMPRS-IS scholar Jenny Schmalfuss won a best paper award at the ECCV Workshop on Adversarial Robustness in the Real World (AROW).
Jenny Schmalfuss joined IMPRS-IS in March of 2021 as a doctoral researcher at the Computer Vision Department of the University of Stuttgart under the supervision of Dr. Andrés Bruhn.
Her research interests are at the intersection of computer vision and machine learning. Here, she researches new ways to quantify and improve the robustness of optical flow methods with the help of adversarial attacks. According to Jenny, "My goal is to enable robust motion estimation by understanding what influences the robustness of current methods."
Last October, Jenny won a best paper award at the ECCV Workshop on Adversarial Robustness in the Real World (AROW). The work is about adversarial attacks with optimized snowflakes, that look and move realistic but can change the predicted motion of typical motion-estimation algorithms into a wrong direction. (https://eccv22-arow.github.io/ and https://eccv22-arow.github.io/short_paper/0014.pdf)
"A Perturbation-Constrained Adversarial Attack for Evaluating the Robustness of Optical Flow" presented by Jenny Schmalfuss at ECCV 2022. Photo credit: University of Stuttgart
Additionally, Jenny more recently co-authored a paper with Lukas Mehl and fellow IMPRS-IS scholar Azin Jahedi which was accepted to CVPR 2023. The paper introduces a new high-detail and high-resolution dataset & benchmark for scene flow, optical flow and stereo. About the paper she states, "Particularly, the benchmark is pretty cool, because people can submit their methods and get the evaluation on this brand-new dataset." The paper can currently be found in preprint (https://www.spring-benchmark.org/ and https://arxiv.org/pdf/2303.01943.pdf).
In addition to her doctoral studies, Jenny is a valuable member of the IMPRS-IS community, volunteering her time as one of the program's S4 student organizers. As a member of S4, she works to create the series of professional development workshops available to our IMPRS-IS scholars. She has also been a featured speaker of the IMPRS-IS Think & Link talk series.
Jenny holds a MSc and BSc degree in Simulation Technology from the University of Stuttgart, where she focused on numerical mathematics, computer vision and machine learning. During her studies, she also worked as a research intern at the National University of Singapore and the University of Houston.
We commend Jenny for the great research she has been doing and thank her for her valuable contributions to our IMPRS-IS doctoral program.
We are proud to have Jenny Schmalfuss recognized as an IMPRS-IS Scholar Spotlight.