Vehicle Classification AI Demo

Lattice sensAI Demo

This demonstration illustrates the use of Lattice Convolutional Neural Network (CNN) Accelerator IP to implement classification of vehicles into cars, trucks and motorcycles/bicycles using Lattice ECP5 devices.

Targeted to the Lattice Embedded Vision Development Kit, one of the onboard cameras provides the input image. The processed image is output through the HDMI output for display. A bounding box is drawn around a vehicle if present. The color of the bounding box is used to indicate the class of the vehicle.

This demonstration was implemented by Lattice sensAI Design Services partner MosChip Technologies.

Features

  • Configuration files provided for standalone operation on the Embedded Vision Development Kit
  • Uses YOLO style network with 224 x 224 input size
  • ECP5 power 0.85W
Lattice sensAIMosChip Logo

Video

Lattice sensAI
Expand Video

Watch a video about Vehicle Classification using ECP5

Block Diagram

Vehicle Classification Demonstration Block Diagrm

Documentation

Quick Reference
Downloads
TITLE NUMBER VERSION DATE FORMAT SIZE
Select All
EVDK Based Vehicle Classification Demonstration User Guide
1.1 6/29/2021 PDF 1.3 MB
TITLE NUMBER VERSION DATE FORMAT SIZE
Select All
EVDK Based Vehicle Classification Demonstration Bitstreams
1.0 9/26/2018 ZIP 18.1 MB

*By clicking on the "Notify Me of Changes" button, you agree to receive notifications on changes to the document(s) you selected.