Dr.-Ing. Bastian Wandt
Bastian Wandt left the Institut für Informationsverabeitung.
Publications and research activities from the time after the departure are not listed here.


  • December, 2020. I was awarded the KI Talent (AI talents) for my dissertation Human Pose Estimation from Monocular Images which is endowed with a price money of 4000€.
  • August, 2020. Our paper Same same but DifferNet: Semi-Supervised Defect Detection with Normalizing Flows got accepted to WACV 2020.
  • July, 2020. Our paper Weakly-supervised Learning of Human Dynamics got accepted to ECCV 2020. The Human Dynamics Dataset will soon be available.
  • December 3, 2019. I'm proud to announce that my student Tom Wehrbein received the KI Talente (AI talents) Award for his bachelor thesis Representation of Human Motion using Neural Networks.


Bastian Wandt studied Mechatronics at the Leibniz Universität Hannover. During his bachelor studies he focused on robotics and automation. He received his B. Sc. in 2012 with a thesis on path planning of autonomous mobile robots. His master thesis "Ganganalyse in monokularen Bildsequenzen mittels eines morphable Shape-Motion-Models" dealt with 3D reconstruction and animation of human motions.

Since 2014 he is working as a research associate at the Institut für Informationsverarbeitung (TNT) and received his Dr.-Ing. in 2020 for the thesis "Human Pose Estimation from Monocular Images". His main research interests are human motion capture and 3D reconstruction. A special focus lies on the application of various machine learning techniques such as dimensionality reduction, compressed sensing, and deep learning. His current research analyses deep neural networks for human pose reconstruction.


Other activities

Supervised theses

  • Klassifikation menschlicher Bewegungen in Unterräumen (2017)
  • Modellierung eines Posenraums menschlicher Bewegungen mit Hilfe neuronaler Netze (2017)
  • Representation of Human Motion using Neural Networks (2018)
  • Codierung menschlicher Oberflächenmodelle mittels Neuronaler Netze (2018)
  • Zeitkonsistente Schätzung Menschlicher Posen mittels Neuronaler Netze (2018)
  • Recurrent Neural Networks for Monocular Human Motion Capture (2019)
  • Realtime 3D Human Pose Estimation (2019)
  • Texturesynthesis using Generative Adversarial Networks (2019)
  • Dimensionsreduktionsmethoden zur Kodierung menschlicher Posen (2019)
  • Tiefenschätzung zur 3D Rekonstruktion menschlicher Posen (2019)
  • Roboterkinematikbasierte Registrierung eines Augmented Reality Systems (2019)