Understanding task performance and crew behavioral health is crucial to mission success and to the optimal design, development, and operation of next-generation spacecraft. Onboard resources, like a conventional 2D video camera, can capture crew motion and interaction; however, there is a critical need for a software tool that achieves unobtrusive, non-invasive, automatic analysis of crew activity from this footage.
The proposed automatic video-based motion analysis software (AVIMA) supports this R&D effort by automatically processing and analyzing complex human motions in conventional 2D video without the use of specialized markers. Unlike many video analytics solutions, AVIMA goes beyond simple blob-based video analysis by tracking the geometric configuration of human body parts like the trunk, head, and limbs. This tracking enables human motion understanding algorithms to model and recognize complex human actions and interactions.
The resulting system will represent a substantial breakthrough providing benefits to an array of applications in video surveillance, human-computer interaction, human factors engineering, and robotics.
Aims for Phase II:
Phase II work will leverage the Phase I preliminary designs and prototype results to design and implement a complete human motion analysis and visualization tool.
1. Enhance the human motion analysis module to automatically estimate 3D human body pose from 2D video footage and to enable quick and robust tracking across a variety of operating scenarios.
2. Investigate and implement a video analytics module that employs state-of-the-art machine learning algorithms to interpret the subjects’ movements with respect to their operational environment, objects in the environment, and other crewmates.
3. Design and implement a user friendly graphical user interface that allows the user to define events/actions to recognize, visualize, and quantify results more quickly.
4. Evaluate the usability, accuracy and robustness of the individual algorithms as well as the end-to-end software system both qualitatively and quantitatively. Test the prototype with potential end-users at NASA.
Aims for Phase II E: Phase II E continues the work from Phase II, focusing on pursuing 2-D video analysis focusing on microgravity video (i.e., floating postures for person detection and tracking)
POTENTIAL NASA COMMERCIAL APPLICATIONS: Vecna expects the full-scope software system to have immediate and tangible benefit for NASA's Exploration Systems Mission Directorate (ESMD). ESMD focuses on the human element of exploration by conducting research to ensure astronaut explorers are safe, healthy and can perform their work during long-duration space exploration. Task performance and crew behavioral health are key concerns in the design, development, and operation of next generation space vehicles. Operations in confined, isolated, and resource-constrained environments can lead to suboptimal human performance. As such, there is a critical need for Vecna's proposed software tool that automatically processes and analyzes crew motion and interaction from video footage captured by a single conventional 2D video camera. Such a diagnostic tool will enable unobtrusive and non-invasive measurement of task performance and crew behavioral health. |