Accelerating Industrial Vision applications from problem to solution. It does not matter if you are beginner or advanced software developer, working for years with OpenCV or just interested to know-how to translate Convolutional Neural Network into a running application.
Join us and learn how to effectively accelerate and reduce time to market of your high performance vision AI application using the Vitis™ unified programming paradigm from Xilinx.
Computer Vision and Image Processing are ubiquitous today in a wide range of applications like Robotics, IIoT, Surveillance, Medical Imaging, ADAS, and Video Streaming services and are also a critical part of the end-to-end processing pipeline of AI-powered vision solutions. During this series you will learn how Xilinx can be used to successfully implement and accelerate your application using Vitis™ Unified Software Platform.From the Neural Network Model to the final inference: AI/ML acceleration within popular software development platforms like Tersor Flow and PyTorch as well as hardware-accelerated open source libraries, Python, C, C++ Support, Apache TVM, pre-trained CNNs – all steps and available tools within Vitis™ will be explained and presented with real examples.
- Session 1: Smart Vision Market Opportunities, Applications, Challenges
- Session 2: Smart Vision Software Application Development and Acceleration
- Session 3: Smart Vision Live Demo using Vitis and Ultra96V2
Session 1 take-aways:
- Where the market opportunities are for smart vision applications
- Key application areas for smart vision
- How to accelerate your Smart vision Application
- How to overcome performance obstacles when developing smart vision applications
Session 2 take-aways:
- How to use standard ML frameworks to implement accelerated smart vision applications
- Which software components allow you to accelerate smart vision applications more effectively
- How to translate your Neural Network Model and run it on the real hardware
- That even without knowing hardware description language, you can use powerful adaptive hardware platform
Session 3 take-aways:
- How to build your first smart vision application using Xilinx About Tool/IP/Demo/Reference
- Designs available now for evaluation and development
- How to step-by-step implement and run the pre-trained Neural Network model
Marketing Lead for Industrial Vision
Field Application Engineer