Addressing System-Level Optimization with OpenVX Graphs, Erik Rainey, Jesse Villarreal, Goksel Dedeoglu, Kari Pulli, Thierry Lepley, Frank Brill, Proceedings of the 2014 IEEE Embedded Vision Workshop (June 2014)
Paper
Abstract During the performance optimization of a computer vision system, developers frequently run into platform-level inefficiencies and bottlenecks that can not be addressed by traditional methods. OpenVX is designed to address such system-level issues by means of a graph-based computation model. This approach differs from the traditional accelera- tion of one-off functions, and exposes optimization possibilities that might not be available or obvious with traditional computer vision libraries such as OpenCV.
Paper
Abstract During the performance optimization of a computer vision system, developers frequently run into platform-level inefficiencies and bottlenecks that can not be addressed by traditional methods. OpenVX is designed to address such system-level issues by means of a graph-based computation model. This approach differs from the traditional accelera- tion of one-off functions, and exposes optimization possibilities that might not be available or obvious with traditional computer vision libraries such as OpenCV.