Amazon and Max Planck Society (also known as Max-Planck-Gesellschaft or MPG) announced the formation of a science node in May 2022. The collaboration marked the first Amazon Science Hub outside the United States. Huben’s goal is to promote research into AI, computer vision and machine learning, the benefits of which are widely shared across all social sectors.
In accordance with this goal, Science Hub sponsors the following research projects:
“Interpersonal Non-verbal synchronous analysis tool (Nisa-Online)”: Senya Polikovsky, Research Engineer, Max Planck Institute for Intelligent Systems
“Interpersonal synchroni (in synchronization), the spontaneous synchronization of behavioral, emotional and even physiological responsibility in contexts with social interaction, provides a useful insight into the quality and smoothness of interactions between social partners. Automatic assessment of Sync could provide objective tools to study the interactive abilities of people with psychiatric states and the degree of therapeutic alliance in psychotherapy as well as online human-to-human and avatar-to-human interactions. Body movement dynamics are central to the analysis of synchronization. In addition, facial expressions, gauze interaction, audio signals and physiological goals such as electrodermal activity, heart rhythm variability or breathing rats may be to detect synchronization. The goal of our proposed research is to expand synchronized analysis to online scenarios. We would like to identify the instruments and procedures for detection of sync in online therapy sessions. “
“3D body shape and clothing estimate from a single image”: Gerard Pons-Moll, Senior Scientist, Max Planck Institute for Informatics and Professor, University of Tübingen, equipped by the Carl Zeiss Foundation
“This research aims to REPTACT A 3D-digital person from a single RGB photograph using a combination of 3D-neural implicit features to absorb worn clothes and a parametric model to absorb and control the posture and shape of the underlying organ. An important news is to incorporate feedback into the deep body estimate network, ensuring that the 3D body network is consisting of its projection in the picture.
“Learning clothing pressure fields via analysis of high faith final element”: Gokhan Serhat, researcher, Max Planck Institute for Intelligent Systems and Deputy, Ku Leuven
“Raising estimates of the clothing pressure on the human body is crucial to understanding physical factors affecting clothing comfort. Such an estimate requires the ability to express contact pressure fields with regard to different body sizes and clothing sizes and material properties. However, such pressure fields are generally too complex to become modulated by analysis methods. Experimentation also has certain disadvantages as it does not provide information about internal tension, and precise measurement of the contact pressure between a few curved soft bodies is difficult. The Instalitic Anatomy of the Human Body may require detailed final elements with thousands of elements for capturing the deformation mechanic’s account. Such large models induce inherently high calculation costs and are likely to suffer from the above numerical. The proposed study aims to tackle this problem by developing a deformation model based on deep neural networks trained with the data from final elemental analysis.
“Material Estimate of Clothing for Controllable Human Avatars”: Justus Thies, research group leader, Max Planck Institute for Intelligent Systems and Professor, Technical University Darmstadt
“An important challenge for controllable human avatars is the reconstruction and animation of clothing. While data -driven, personal -specific restructuring methods show promising results, the extrapolation of new poses with realistic deformations of clothing is limited. Unlike data-driven methods, physics simulation may be to model the dynamic changing clothing. In this project we want to investigate material estimate of human clothing for such simulations. Specifically, we are interested in predicting Matti’s speedies such as stiffness or stretching that affect the movement’s movement depends. In addition to these mechanical material properties, we are also interested in predicting material properties for the appearance of clothing that allow for re -releases during new poses.