Personal information

computer vision, machine learning, human pose estimation, deep learning, visual person analysis
Germany, Hungary

Biography

István Sárándi is a postdoctoral researcher at the University of Tübingen (Germany) in the Real Virtual Humans group headed by Prof. Dr. Gerard Pons-Moll. He received the M.Sc. and PhD degrees in computer science from RWTH Aachen University (Germany), and the B.Sc. degree in computer engineering from the Budapest University of Technology and Economics (Hungary). His research focuses on visual 3D human analysis. Methods from his first-author publications achieved First Place both in the 2018 ECCV PoseTrack Challenge on 3D human pose estimation and in the 2020 ECCV 3D Poses in the Wild Challenge.

Activities

Employment (2)

University of Tübingen: Tübingen, Baden-Württemberg, DE

2023-03-01 to present | Postdoctoral Researcher (Real Virtual Humans group)
Employment
Source: Self-asserted source
István Sárándi

RWTH Aachen University: Aachen, Nordrhein-Westfalen, DE

2017-04-03 to 2023-02-28 | Research Assistant & PhD Student (Computer Vision Group)
Employment
Source: Self-asserted source
István Sárándi

Education and qualifications (3)

RWTH Aachen University: Aachen, Nordrhein-Westfalen, DE

2017-04-03 to 2023-02-28 | Dr. rer. nat. (Computer Vision Group)
Education
Source: Self-asserted source
István Sárándi

RWTH Aachen University: Aachen, Nordrhein-Westfalen, DE

2012-10 to 2016-03 | Computer Science MSc
Education
Source: Self-asserted source
István Sárándi

Budapest University of Technology and Economics: Budapest, HU

2008-09 to 2012-01 | Computer Engineering BSc
Education
Source: Self-asserted source
István Sárándi

Works (12)

Learning 3D Human Pose Estimation from Dozens of Datasets using a Geometry-Aware Autoencoder to Bridge Between Skeleton Formats

IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)
2023-01 | Conference paper
Contributors: István Sárándi; Alexander Hermans; Bastian Leibe
Source: Self-asserted source
István Sárándi

MeTRAbs: Metric-Scale Truncation-Robust Heatmaps for Absolute 3D Human Pose Estimation

IEEE Transactions on Biometrics, Behavior, and Identity Science
2020 | Journal article
Part of ISSN: 2637-6407
Version of ARXIV: 2007.07227
Source: Self-asserted source
István Sárándi
grade
Preferred source (of 2)‎

Reposing Humans by Warping 3D Features

2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
2020-06 | Conference paper
Part of ISBN: 9781728193601
Version of ARXIV: 2006.04898
Source: Self-asserted source
István Sárándi

Metric-Scale Truncation-Robust Heatmaps for 3D Human Pose Estimation

2020 15th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2020) (FG)
2020-05 | Conference paper
Version of ARXIV: 2003.02953
Source: Self-asserted source
István Sárándi
grade
Preferred source (of 2)‎

Visual Person Understanding Through Multi-task and Multi-dataset Learning

German Conference on Pattern Recognition (GCPR)
2019 | Conference paper
Part of ISBN: 9783030336752
Part of ISBN: 9783030336769
Part of ISSN: 0302-9743
Part of ISSN: 1611-3349
Version of ARXIV: 1906.03019
Source: Self-asserted source
István Sárándi

Synthetic Occlusion Augmentation with Volumetric Heatmaps for the 2018 ECCV PoseTrack Challenge on 3D Human Pose Estimation

2018-09-13 | Preprint
Source: Self-asserted source
István Sárándi

How Robust is 3D Human Pose Estimation to Occlusion?

2018-08-28 | Preprint
Source: Self-asserted source
István Sárándi

Quantitative Conjunctival Provocation Test for Controlled Clinical Trials

Methods of Information in Medicine
2014 | Journal article
Part of ISSN: 0026-1270
Part of ISSN: 2511-705X
Source: Self-asserted source
István Sárándi

Towards quantitative assessment of calciphylaxis

Medical Imaging 2014: Computer-Aided Diagnosis
2014 | Conference paper
Source: Self-asserted source
István Sárándi

Zuordnung von Nationalem Kompetenzbasiertem Lernzielkatalog und fakultätsspezifischen Lernzielkatalogen durch automatische Textverarbeitung

2014 | Conference poster
Source: Self-asserted source
István Sárándi

Automatic conjunctival provocation test combining Hough circle transform and self-calibrated color measurements

Medical Imaging 2013: Computer-Aided Diagnosis
2013 | Conference paper
Source: Self-asserted source
István Sárándi

Quantitative Conjunctival Provocation Test

GMDS 2013; 58. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik
2013 | Conference abstract
Source: Self-asserted source
István Sárándi

Peer review (6 reviews for 5 publications/grants)

Review activity for IEEE multimedia. (1)
Review activity for IEEE transactions on biometrics, behavior, and identity science. (1)
Review activity for IEEE transactions on neural networks and learning systems. (1)
Review activity for IEEE transactions on pattern analysis and machine intelligence. (2)
Review activity for The visual computer (1)