AI-powered image dataset generator goal of VISIMO Air Force contractNews
December 21, 2021
PITTSBURGH, Pa. VISIMO, with Research Institute partner Colorado State University, has won an Air Force Phase II Small Business Technology Transfer (STTR) award to develop a generator capable of quickly creating synthetic, annotated image datasets.
According to the company, Air Force Research Laboratory, Information Directorate (AFRL/RI) is collaborating as the end-user of VISIMO's technology. AFRL/RI works to prototype innovative technologies with hopes to transition them to potential users across the Air Force and Department of Defense.
Researchers are often forced to manually create training datasets, which officials claim can be tedious. Models that train on overhead imagery include tracking and detection algorithms, which are the types of models used for autonomous vehicle navigation, humanitarian aid drops, wildfire management, search and rescue, and others.
VISIMO is building a conditional generative adversarial network (CGAN) that learns to generate unlimited original backgrounds from a limited amount of landscape images. In Phase I, the CGAN proof-of-concept focused on overhead satellite imagery and in Phase II it will be expanded to other types of image data, like SAR or radar/lidar.