CIVIT Tech Day: Motion Deconstructed on YouTube

You can find all the presentations from the CIVIT Technology Day: Motion Deconstructed now on YouTube. The event was organized at Tampere University on Friday 26 April 2019. 


“Tell Me How You Look and I’ll Tell You How You Move” by Thiemo Alldieck, Technical University of Braunschweig, Germany

Kinematics of Music-Induced Movement 
Petri Toiviainen, Finnish Centre for Interdisciplinary Music Research, University of Jyväskylä, Finland

Motion Capture with Optitrack: A live tutorial
by Olli Suominen, CIVIT / Tampere University, Finland

The Role of Gyroscopes in Motion Sensing
by Jussi CollinJC Inertial Oy / Tampere University, Finland

Motion Sensing for Excavators
by Antti KoluNovatron

Virtual Studio Solutions
by Mikko KarsistoKeho Interactive

Presentations and Speakers
Tell Me How You Look and I'll Tell You How You Move

Human body shape and motion estimation are two sides of the same coin. To be able to fully understand human motion from monocular imagery, we need to understand the shapes of the tracked subjects, too. In my talk, I will motivate why we need better shapes for better tracking. I will demonstrate how 3D bodys helped us to understand human motion better and where these models find their limits. If we want to learn rich models of human shape, motion, and dynamics, we require new approaches that learn from ubiquitous data such as plain RGB-images and video. I continue with discussing recent advances in personalized body shape estimation from monocular video, from a few frames, and even from a single image. We developed effective methods to learn detailed avatars without the need for expensive scanning equipment. These methods are easy to use and enable various VR and AR applications. I will conclude my talk by outlining the next challenges in human shape reconstruction and how this potentially affects human motion estimation.

Thiemo Alldieck obtained his Master’s degree from Aalborg University in Vision, Graphics and Interactive Systems in 2015. Since 2016 he is a Ph.D. student at the Computer Graphics Lab at TU Braunschweig. In 2018 he was a visiting Ph.D. in the Real Virtual Humans and Graphics Vision & Video groups at Max-Planck Institute for Informatics, Saarbrücken. He works in close cooperation with the groups on the topic of monocular human shape reconstruction. His work has been published in various computer vision conferences including CVPR, 3DV, and GCPR.

Kinematics of Music-Induced Movement

Music is commonly regarded as being primarily an auditory phenomenon, and the bulk of research on music perception focuses exclusively on this sensory modality. Only recently have the multimodal aspects of music processing gained significant attention. In addition to the auditory, other sensory modalities such as visual, tactile, and proprioceptive, play a significant role in the processing of musical information. A common feature of the latter modalities is that they rely on corporeal movement.

Examples of the important role of movement in music are abundant. For instance, in most cultures music is associated with dance. Furthermore, we tend to move while listening to music, in particular when the music has a clear pulse. Additionally, corporeal movement is important for the interaction and communication between musicians, helping to maintain synchronization and convey expressive intentions. Lastly, movement is an important ingredient in social interaction associated with music listening. Music can thus be regarded as a fundamentally spatiotemporal phenomenon.

In my talk I will review work on music-induced movement carried out at our lab. Specific topics include kinematics of synchronization to musical beat, influence of musical content and personality on music-induced movement as well as kinematics of movement interaction in spontaneous dance.

Petri Toiviainen received the degree of PhD in musicology in 1996 at the University of Jyväskylä, Finland. Since 2002 he holds the position of Professor of Music at the University of Jyväskylä. He has been visiting professor at Cornell University and visiting fellow at the Center for Advanced Study in the Behavioral Sciences at Stanford University. During 2008-13, he was the head of the Finnish Centre of Excellence in Interdisciplinary Music Research, located at the universities of Jyväskylä and Helsinki. In 2014-18 he held an Academy Professorship granted by the Academy of Finland. His research interests include music processing in the brain, music and movement, sound and music computing, and music visualization.

Motion Capture with Optitrack: A live tutorial
Naturalpoint Optitrack is one of the world leading optical motion capture systems used in the cinema and gaming industries to capture movement from real actors and transforming them to realistic animations on virtual characters. Aside from creating animations, the same system can also be used to track anything in sub-millimeter accuracy in spaces in the scale of 10×10 meters. CIVIT is equipped with a 16 camera Optitrack motion capture system, which can also be deployed elsewhere with relative ease. This presentation will give an overview to the operating principle of the system, show how it is setup and used to track different objects, and how to access the tracked data.
Olli Suominen graduated with both BSc and MSc (Tech) in Information Technology from Tampere University of Technology (2011/2012) with a major in Signal Processing. Since then, he has been a PhD student in 3D Media Group at the Laboratory of Signal Processing at TUT and managing the construction and development of CIVIT. After starting from the BSc thesis using only one camera (Depth Image Based Rendering) and an MSc thesis using two cameras (Stereo Depth Estimation) he has now scaled up to 40 cameras for the PhD with research interests in multi-camera systems, 3D reconstruction, multimodal sensor fusion, SLAM and light field capture. He currently focuses on applications in heavy mobile work machines, leading several industry driven research projects and developing relations with the industry.

The Role of Gyroscopes in Motion Sensing
Auto-rotating the phone screen according to orientation is a well-known example of accelerometer application. If one shakes the phone the image shown may be in wrong orientation. Similar problem may be encountered if one tries to use a bubble level in a moving vehicle. Gyroscopes are used to avoid such problems. In this presentation, we take a look what accelerometers and gyroscopes really measure and show how to fuse their information to obtain relevant information on motion, orientation and vibration.
Jussi Collin received the M.Sc. and Dr.Tech. degrees from the Tampere University of Technology, Tampere, Finland, in 2001 and 2006, respectively, specializing in applications of inertial sensors. He is currently CEO of JC Inertial Oy and adjunct professor at Tampere University. JC Inertial Oy is developing HW and SW that gets the most out of MEMS inertial sensors. By combining motion sensing devices with modern GNSS receivers an extraordinary accuracy with unprecedented availability is obtained.

Motion sensing for excavators

Digitalization and automation are changing in the field of earthmoving industry. Novatron has been developing machine control systems for mobile machinery for over 25 years. With Novatron’s machine control systems we can track the excavators movements and visualize them together with Buildin Information Models(BIM). I will talk about the benefits this brings to our customers and how it is done. I will also present a glimpse into the future of machine control.

Antti Kolu graduated with MSc (Tech) in Automation and Information Technology from Tampere University of Technology 2010. He continued as a Researcher and Doctoral student at the department of Automation and Hydraulic Engineering(AUT) where his research concentrated on surround sensing, object detection and path planning for intelligent mobile work machines. Has taken part in many academic and industry driven projects. In 2018 he started to work at Novatron as a Motion Planning Scientist where he develops new innovations and world leading solutions for earthmoving automation.

Virtual Studio Solutions

The role of XR in television, as seen through Keho Interactive‘s projects. What will the future hold? What are the breakthrough advances in technology, that we expect in the coming years? Keho Interactive is a Tampere-based company that develops technologies and methods for combining the real and the virtual. The company offers motion capture and other motion-based digital services for television, game and movie industries, sports and performing arts. Keho Interactive’ virtual studio solution is the most flexible Virtual and Augmented Reality studio solution for broadcasting and streaming.

Mikko Karsisto, CEO, Keho Interactive Ltd. Keho Interactive Ltd develops and operates one of the most advanced virtual studios in the industry. In addition, it offers animation and digital performance capture services for games, movies and television and other motion-based digital services.

CIVIT Tech Day: Motion Deconstructed on YouTube

Close Menu