fusion was built to support deeply embedded System on Module and System in Package chipsets, deployment on some cameras, edge, fog, cloud, and server based computing environments. fusion’s process and workflow design tooling requires a Windows powered PC. fusion executable solution kits and algorithms run on Windows operating systems as well as linux based operating systems like ubuntu and android.
fusion is an open and extensible integrated development environment and platform that developers and analysts can use to create simple to complex algorithms, workflows, and solution kits. Users may choose from a variety of proprietary and open source machine learning and object detection models including TensorFlow, YOLO v3, and more including proprietary object detection and machine learning models built by and owned by our development team. In addition, fusion provides users with hundreds of proprietary capabilities, functions, algorithms, activities detectors, filters, and more that complement OpenVINO as well as other third party contributions. Altogether, fusion provides our partners and clients with the easiest, fastest, and highest level of abstraction and comprehensiveness that dramatically improves the speed of development and the quality, precision, and accuracy of the video and data analytics solutions you create.
fusion includes a robust library of components and functionalities that are easily applied by the solution designer that provide pre-processing of the video and post-processing of the data that is necessary to create a robust, precise, and comprehensive solution. The pre-processing and post-processing are critical steps in our algorithms and solutions that make a significant difference between detection and recognition and provide a refined and efficient mechanism that delivers precise, accurate, and reliable results. These pre- and post-processing steps include advanced statistics, data filters, applied mathematics, predictive modeling and control, and other analytics to offer nuanced and trusted understandings, context, relationships, and insights of various use cases and complexities.
fusion includes proprietary deep learning and inferencing convolution neural networks. fusion also supports third party open source detectors and inferencing engines like tensorflow and YOLOv3 that are included in the fusion platform.
fusion algorithms and executables may be deployed on Intel CPU, GPU, VPU, or FPGAs as well as NVIDIA GPUs and SOMs and ARM processors.
the fusion platform leverages .Net Core 5. this gives fusion cross platform compatibility with Xamarin Android, Xamarin IOS, Windows UWP, and Linux.
add new capabilities to fusion that you own and include them in uniquely yours solution kits.