Efficient Large-Scale CNN-enabled Semantic Segmentation of Video
Abstract
The recipient will develop the content-based image retrieval (CBIR) area of the Command, Control, Communications, Computers, Intelligence, Surveillance and Reconnaissance (C4ISR) Enterprise to the Edge (CETE) program. A transformative capacity for integrated visual perception including classification, detection, and segmentation driven by advanced methods for deep learning realized in an efficient algorithmic and software framework. The recipient also plans to scale convolutional neural networks (CNN)-based models to provide efficient semantic segmentation with both large-scale label spaces and jointly-optimized and learned fine-grained pixel-level assignment maps.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 11, 2016
- Source ID
- FA87501510091
Entities
People
- Trevor Darrell
Organizations
- Rome Laboratory
- United States Air Force
- University of California Regents