Hardware components | ||||||
 |
| × | 1 | |||
Software apps and online services | ||||||
| ||||||
 |
| |||||
Overview
i use the Vitis to do the development.
Vitis AI itself is composed of :
- AI Model Zoo - Pre-optimized models that are ready to deploy on Xilinx devices including Kria
- AI Optimizer - An optional model optimizer that can prune a model by up to 90%. It is separately available with commercial licenses.
- AI Quantizer - A powerful quantizer that supports model quantization, calibration, and fine tuning.
- AI Compiler - Compiles the quantized model to a high-efficient instruction set and data flow.
- AI Profiler - Perform an in-depth analysis of the efficiency and utilization of AI inference implementation.
- AI Library - C++ APIs for AI applications from edge to cloud. i use this to send data to cloud
DPU - Efficient and scalable IP cores can be customized to meet the needs for many different applications.
- For more details on the different DPUs available, refer to DPU Naming.
- DPU - Efficient and scalable IP cores can be customized to meet the needs for many different applications.
For more details on the different DPUs available, refer to DPU Naming.
Vitis AI (xilinx.com) describe details about Vitis
Setting Up
Follow the Getting Started steps listed on GitHub - Xilinx/Vitis-AI: Vitis AI is Xilinx’s development stack for AI inference on Xilinx hardware platforms, including both edge devices and Alveo cards.
Vitis Video Analytics SDK
Vitis Video Analytics SDK (xilinx.com) supports video and RTSP
The Vitis Video Analytics SDK is an optimized graph architecture built using the open-source GStreamer framework. The graph below shows a typical video analytic application starting from input video to output metadata
Model Zoo
Xilinx has made available many pre trained models known AI model zoo
The list of models that i plan to try is SSD and Yolov3. The full list of models can be found at https://github.com/Xilinx/Vitis-AI/tree/master/models/AI-Model-Zoo/model-list
AI Library
Test
the rtsp
Running the model
without further training these are the result
Comments