By utilizing its development tools (eCV SDK and DeepView ML Toolkit), Au-Zone enables its customers to quickly develop and securely deploy machine learning solutions and novel Convolutional Neural Networks on hardware-accelerated, heterogeneous computing devices.
As a Toradex ecosystem partner, Au-Zone helps clients lower development costs, mitigate program risk and shorten time-to-revenue when designing new vision-enabled products.
Machine Learning, Machine Vision, OS & Driver Development
DeepView is a graphical Toolkit and hardware accelerated Run-Time inference engine specifically developed to help embedded engineers design, train and deploy Convolutional Neural Networks (CNNs) on embedded platforms. With desktop support for Windows, Mac and Linux, the Toolkit enables developers to create and curate datasets, design custom networks or select from existing templates and train with a single click using the TensorFlow framework from Google. Once satisfied with the network design on the desktop, the model can be validated, optimized and deployed directly on the embedded target using the DeepViewRT, the Run-Time Inference Engine.
DeepViewRT, the runtime Inference Engine is supported across Linux, RTOS and bare metal environments and supports a wide range of heterogeneous compute and vision accelerated devices.