September 28, 2012
Teams live and work together to explore new technologies over 8-weeks
What happens when six teams of diverse, creative innovators are chosen to live and work together to try creative and ambitious approaches to hard technical challenges facing the Department of Defense? DARPA’s Innovation House Study, an experiment in collaboration and research that kicked off its 8-week sprint last week at George Mason University’s Arlington, Va., campus, is about to find out.
The program provides a focused residential research environment for six teams in the area of extracting meaningful content from large volumes of varied visual and geospatial media. The selected teams receive $50,000 to cover expenses and time over the 8 weeks.
“This experiment allows thinkers from diverse fields to zero in on focused area of research for eight weeks,” said Mike Geertsen, DARPA program manager and the force behind the Innovation House Study. “We are examining how collaboration among different disciplines can yield game-changing technologies. For instance, our participants include computer engineers, data architects and imagery analysts. We have thrown them in with a robotics expert from NASA, visual artists and neuroscientists. If this model proves to be successful, it represents a new means of tackling some of the hard problems in government-sponsored research.”
At Tuesday’s innovation House kick-off, the teams gave short presentations on their proposed areas of study:
Petavision
Multi-Modal Approaches to Real-Time Video Analysis
Biologically-inspired, hierarchical neural networks to detect objects of interest in streaming video by combining texture/color, shape and motion/depth cues.
GOSE (Geospatial-Oriented Structure Extraction)
Structure Extraction from Imagery
Automatic construction of a 3D wireframe of an object using as few images as possible from a variety of angles.
Videovor
Visualization of Video Information
Software that fragments and re-models linear time of video content to cluster big pools of related data in a 3D interactive interface for human analysis.
Hybridcv
Inferring Objects in Photos and Video
A system that allows the user to infer the location and structure of hidden object parts from photos and videos as well as provide categorization and recognition via an associative memory framework.
Looking Glass
Focused Sensor Fusion
A new integrated working environment for analysts that presents relevant data from a large variety of geospatial data and sensors, including ground-level video, LiDAR, wide-area airborne full motion video, and unstructured data.
IVASIA
Visualization for Image Analysis
A lightweight virtual world in 3D perspective that would follow the work that an image analyst is performing.
Teams are expected to produce an initial design and demonstrate in software the crucial capabilities that validate their approach in the first four weeks of the project. After which, teams are expected to complete and demonstrate a functional software configuration as a proof of concept.
DARPA is providing the teams with access to data and facilitating ongoing interaction with mentors from the U.S. Government and academia. These interactions will provide teams with context for how their proposed technology could be used by the warfighter or analyst.
Details on the program can be found at: http://c4i.gmu.edu/InnovationHouse/.
###
Associated images posted on www.darpa.mil may be reused according to the terms of the DARPA User Agreement, available here: http://go.usa.gov/nYr.
Tweet: @darpa
Hashtag: #innovation