István Sárándi
I'm a PhD student in Computer Vision and Machine Learning, advised by Bastian Leibe in the Computer Vision Group of RWTH Aachen University in Germany.
My research focus is 3D human understanding with robotics application scenarios in mind.
BSc Computer Engineering
Budapest Univ. of Tech. and Eco., Hungary
MSc Computer Science
RWTH Aachen University, Germany
PhD Computer Science
RWTH Aachen University, Germany

Before joining as a PhD student, I worked in the same group as a master student on crowd counting from images and video. Before that, I designed and analyzed medical image processing algorithms at the Institute of Medical Informatics at RWTH and published papers on eye redness measurement and wound imaging. During my German-language undergrad at the Budapest University of Technology and Economics in Hungary, I specialized in Autonomous and Intelligent Systems. I wrote my thesis at Hungary's National Institute for Strategic Health Research on machine learning approaches for diagnosis classification. In an exchange semester and an internship at the Karlsruhe Institute of Technology, I focused on robotics, data mining and image processing.


Reposing Humans by Warping 3D Features
Markus Knoche, István Sárándi, Bastian Leibe
Workshop on Human-Centric Image/Video Synthesis, Conference on Computer Vision and Pattern Recognition (CVPRW), 2020.
Metric-Scale Truncation-Robust Heatmaps for 3D Human Pose Estimation
István Sárándi, Timm Linder, Kai O. Arras, Bastian Leibe
IEEE International Conference on Automatic Face and Gesture Recognition (FG), 2020.
Visual Person Understanding through Multi-Task and Multi-Dataset Learning
Kilian Pfeiffer, Alexander Hermans, István Sárándi, Mark Weber, Bastian Leibe
German Conference on Pattern Recognition (GCPR), 2019.
Synthetic Occlusion Augmentation with Volumetric Heatmaps for the 2018 ECCV PoseTrack Challenge on 3D Human Pose Estimation (1st Place)
István Sárándi, Timm Linder, Kai O. Arras, Bastian Leibe
Extended abstract at the ECCV PoseTrack Workshop (ECCVW), 2018.
How Robust is 3D Human Pose Estimation to Occlusion?
István Sárándi, Timm Linder, Kai O. Arras, Bastian Leibe
Workshop on Robotic Co-Workers 4.0 at IEEE/RSJ International Conference on Intelligent Robots and Systems (IROSW), 2018.

Non-Computer Vision Publications and Abstracts

Quantitative Conjunctival Provocation Test for Controlled Clinical Trials
István Sárándi, Dan Philipp Claßen, Anatoli Astvatsatourov, Oliver Pfaar, Ludger Klimek, Ralph Mösges, Thomas M. Deserno
Methods of Information in Medicine, 2014;53(4):238-44
Zuordnung von Nationalem Kompetenzbasiertem Lernzielkatalog und fakultätsspezifischen Lernzielkatalogen durch automatische Textverarbeitung
Cord Spreckelsen, Roland Döpke, István Sárándi
Conference of the German Association for Medical Education (GMA), 2014.
Towards Quantitative Assessment of Calciphylaxis
Thomas M. Deserno, István Sárándi, Abin Jose, Daniel Haak, Stephan Jonas, Paula Specht, Vincent Brandenburg
SPIE Medical Imaging 2014: Computer-Aided Diagnosis (Vol. 9035, p. 90353C), 2014.
Automatic Conjunctival Provocation Test Combining Hough Circle Transform and Self-Calibrated Color Measurements
Suman Raj Bista, István Sárándi, Serkan Dogan, Anatoli Astvatsatourov, Ralph Mösges, Thomas M. Deserno
SPIE Medical Imaging 2013: Computer-Aided Diagnosis (Vol. 8670, p. 86702J), 2013.
Quantitative Conjunctival Provocation Test
István Sárándi, Thomas Deserno, Dan Classen, Oliver Pfaar, Anatoli Astvatsatourov, Ralph Mösges
Conference of the German Association for Medical Informatics, Biometry and Epidemiology (GMDS), 2013.

Theses and Term Papers

Pedestrian Line Counting by Probabilistic Combination of Flow and Appearance Information
István Sárándi
RWTH Aachen University, Germany, 2015.
Feature Mining for Localized Crowd Counting
István Sárándi
Seminar Current Topics in Computer Vision and Machine Learning, RWTH Aachen University, 2014.
Subspace Clustering of High Dimensional and Streaming Data
István Sárándi
Seminar Multimedia Retrieval and Data Mining, RWTH Aachen University, 2013.
Computing Semantic Similarity Between Medical Learning Objectives Across Catalogues
István Sárándi
Institute of Medical Informatics, RWTH Aachen University, 2016.
System Development to Support Medical Coding
(HU: Egészségügyi Kódolástámogató Rendszer Fejlesztése)
István Sárándi
Budapest University of Technology and Economics, Hungary, 2011.