SBIR/STTR Research Award
Building computer vision products for the real world
Award amount
$50,000Program and phase
SBIR · Phase IAward year
2020Award details
- Awarding agency
- Department of Defense
- Branch
- Air Force
- Recipient company
- LABELBOX, INC.
- Proposal award date
- 2020-03-09
- Contract end date
- 2020-06-08
- UEI
- ZEHGNCJX69X6
- Solicitation number
- 20.1
- Topic code
- J201-CSO1
Research abstract
Labelbox is a foundational software platform for the creation and management of high-quality training data for AI. It enables users to label images and video, which can then be used to train AI models. This allows the warfighter to internally build and maintain an AI training data infrastructure--a key enabler of success as AI is integrated throughout the Air Force. Labelbox is not just a labeling tool, our platform also incorporates additional features for collaboration between labelers, data science teams, project management, quality assurance, accuracy improvement, and the “rework” of images to improve upon existing labels. Not only does Labelbox provide the Air Force with a reliable solution for automation, it also assists with reducing man hours and cost associated integration of automatic processes.
Technology keywords
Not reported
Awards with the same topic code
Verified by an exact topic code match. This is not a text-similarity match.
- AI-enabled Test Incident Report Analysis Tool
- Flight Development of an ASCENT Propellant Decomposition Chamber
- Mage Duel Serious Game - TACFI
- Model Řne: Scalable Multi-Domain Unified C3 Architecture via Secure and Federated Models
- FA557024Q0006 - TACFI - PeakMetrics - JICAF Innovation Information Domain Initiative