IPi Motion Capture V2.0
Affordable, markerless mocap technology just got a little more interesting. Moscow-based iPi Soft has released an updated version of its mocap software aimed at television, film and gaming animators and VFX artists that now supports up to 16 USB cameras, like Sony's PlayStation Eye Camera, and up to four Microsoft Kinect 2 sensors. The new software can also track three actors simultaneously, a boon for animators and developers looking to more accurately capture group and team activity such as sports and dance moves and fight sequences.
IPi Motion Capture V2.0
A promising alternative for manual coding is automated coding using motion-tracking technologies, which can track the movements of multiple individuals simultaneously and quantify their interactions. Such technology has been used extensively in the study of social interactions in animals17, enabling large studies of fruit flies18 and mice19, including mouse models of ASD20. Similar approaches have been used to study human behavior in adults21, and even to identify repetitive movements in older children with ASD22 or typically developing adults mimicking such movements23. However, to the best of our knowledge this technology has not been used so far to study social interactions in children or to assess the severity of social symptoms in ASD children.
We examined how each of several motion tracking measures of child-clinician interaction were correlated with the total ADOS scores of the children. We also examined the social affect (SA) and restricted and repetitive behavior (RRB) scores of the ADOS separately, to assess the specific relationship between the motion tracking measures and the severity of distinct ASD symptoms.
The goal of the current study was to demonstrate that motion tracking algorithms can automate some of the measures used to quantify the severity of social symptoms in young ASD children. By focusing on robust measures of distance and quantification of approach and avoidance behaviors, is was possible to explain a considerable part of the variability in ADOS scores across children. Note that the examined motion tracking measures were more strongly related to social symptoms (i.e., correlated with ADOS SA scores), rather than restricted and repetitive behaviors.
In contrast to contemporary behavioral research with ASD children, which is heavily reliant on manual coding techniques, behavioral research with ASD animal models relies heavily on objective, automated behavioral tracking techniques35. Most interesting are ecological motion tracking techniques that can identify multiple individual animals in a social setting and track their behavior relative to each other17. Such systems have been used, for example, to characterize the structure of social interactions in fruit flies18 and large-scale group social dynamics of mice19. A recent study with BTBR mice, a popular ASD mouse model36, used similar motion tracking techniques to demonstrate that BTBR mice remain further away from control mice and exhibit reduced social exploration behaviors20.
A variety of new commercially available technologies for tracking human behaviors are emerging37. While several studies have reported that similar motion tracking techniques can identify postural problems38 and certain repetitive behaviors22 in children with ASD, to the best of our knowledge, this is the first study to implement measures of social interactions involving the relative movements of multiple individuals using motion tracking. The ability to quantify social symptoms in young ASD children in this manner will enable further research to relate specific findings across human and animal studies. Furthermore, combining additional automated measures from other behavioral domains such as speech (e.g., human speechome project39) may offer additional utility in quantifying the development of social symptoms.
Thanks to the abstraction layer that the library provides, it has become very easy to monitor the processor metrics inside your application. Before their usage, the performance counters need to be initialized. Afterwards, the counter state can be captured before and after the code section of interest. Different routines capture the counters for cores, sockets, or the complete system, and store their state in corresponding data structures. Additional routines provide the possibility to compute the metric based on these states. The following code snippet shows an example for their usage:
The Xeon E5 series processor's uncore has multiple 'boxes' similar to the Xeon E7 processor (Intel microarchitecture codename Westmere-EX). Intel PCM v2.0 supports IntelQPI and memory metrics for the new processor.
One of the really strong points of using the iPi Soft Desktop Motion Capture is how clean and simple a program you have to interface with while still allowing users to work with different camera options. Users have a choice of working with as many as 6 Playstation Eyes, 2 Kinects Xbox, or the new ASUS Xtion camera in order to track a motion performance. After the recordings have been imported, iPi Studio not only syncs the video to a rig, it also comes with controls to clean up the tracking data before exporting to animation software. Most all major animation packages are covered for export including Maya, MotionBuilder, 3D Max, Softimage, Blender, Maxon Cinema 4D, and the list goes on.
For example, What if you needed a very specific movement for a project? What if you wanted to motion capture your own movements? Do you need to rent one of those ping-pong ball suits?! I was just as curious as you so I took some time to research and test a DIY motion capture system that can be imported into Cinema 4D. The result is my recreation of the "crane kick" scene from the original Karate Kid movie. I've even setup a free project file for you to download and mess around with. Enjoy!
After doing some research I found a great DIY motion capture rig to be iPi Soft mixed with an Xbox Kinect Camera. The result was even better than I originally imagined.
iPi says you can only record frontways on a single camera. However, I spun around and... oh my goodness, it worked!Bear in mind this is the only software I've tested using this technique. If you use any other applications to test out DIY motion capture please tell us about your experience. I've listed them at the end of this article for reference.
Motion capture is a rabbit hole that can get REALLY deep. If you're looking for some alternative methods to those listed here in this article, here are some different motion capture solutions from around the industry.
iPi Recorder is a software program provided by iPi Soft for capturing, playing back and processing video records from multiple cameras and depth sensors. Captured records can be used for motion tracking in iPi Mocap Studio.iPi Recorder supports the following cameras and depth sensors: