Zurück zur Übersicht Mittwoch, 10.02.2021
Mittwoch | 15.30 Uhr | Scientific Computing meets Neuroscience @ UHEI Unsupervised Behavior Analysis and Magnification using Deep Learning (uBAM) Prof. Dr. Björn Ommer, Heidelberg University, Interdisciplinary Center for Scientific Computing Motor behavior analysis is essential to biomedical research and clinical diagnostics as it provides a non-invasive strategy for identifying motor impairment and its change caused by interventions. State-of-the-art instrumented movement analysis is time- and cost-intensive, since it requires placing physical or virtual markers. Besides the effort required for marking keypoints or annotations necessary for training or finetuning a detector, users need to know the interesting behavior beforehand to provide meaningful keypoints. We introduce uBAM, a novel, automatic deep learning algorithm for behavior analysis by discovering and magnifying deviations. We propose an unsupervised learning of posture and behavior representations that enable an objective behavior comparison across subjects. A generative model with novel disentanglement of appearance and behavior additionally visually magnifies subtle behavior differences across subjects directly in video without requiring a detour via keypoints or annotations. Evaluations on rodents and human patients with neurological diseases demonstrate the wide applicability of our approach. Moreover, combining optogenetic stimulation with our unsupervised behavior analysis shows that our approach can also serve as a noninvasive diagnostic tool for measuring structural plasticity in the cortex. Adresse - - - 0 - Homepage Veranstaltung https://typo.iwr.uni-heidelberg.de/events/iwr-colloquium/ Veranstalter Interdisciplinary Center for Scientific Computing Homepage Veranstalter https://typo.iwr.uni-heidelberg.de/home/ Kontakt Kontakt URL https://www1.iwr.uni-heidelberg.de/iwr/public/people/de/3217 Alle Termine der Veranstaltung 'Scientific Computing meets Neuroscience @ UHEI': The IWR Colloquium serves as a platform for the interdisciplinary dialogue which characterizes the field of scientific computing. Mittwoch, 16. Dezember 2020, 16.30 Uhr From Bistable Neurons to Recurrent Neuronal Networks Dr. Rebecca Mease, Heidelberg University, Medical Faculty Heidelberg, Institute of Physiology and Pathophysiology Mittwoch, 20. Januar 2021, 16.30 Uhr Scientific Computing meets Neuroscience @ UHEI Mittwoch, 03. Februar 2021, 16.30 Uhr Statistical Approach for Parameter Identification of Reaction-Diffusion Models Dr. Alexey Kazarnikov, Heidelberg University, Institute for Applied Mathematics Mittwoch, 10. Februar 2021, 15.30 Uhr Unsupervised Behavior Analysis and Magnification using Deep Learning (uBAM) Prof. Dr. Björn Ommer, Heidelberg University, Interdisciplinary Center for Scientific Computing |