Object Counting

Lattice sensAI Demo

ECP5 based Convolutional Neural Network (CNN) acceleration for Object Counting with 7 convolution layers implemented in 8 Neural Network (NN) engines on ECP5-85 FPGA.

Optimized Neural Network for small devices.

Standalone operation based on Embedded Vision Development Kit counting fruits.

Features

  • Runs @ 1fps with 224 x 224 RGB Input
  • Total ECP5 power consumption of 0.85 W
  • Uses weights and activation based on a fruit dataset created by Lattice
  • Internal EBR blocks used to store activations, minimizing DRAM access
Lattice sensAI

Jump to

Video

Object Counting Using ECP5 and CNNsExpand Image

Object Counting Using ECP5 and CNNs

  • This demonstration tallies apples and oranges to demonstrate object counting
  • The inferencing is done using eight Convolutional Neural Networks implemented in the Embedded Vision Development Kit’s ECP5 FPGA
  • Power consumption is less than 1W

Block Diagram

Documentation

Quick Reference
Downloads
TITLE NUMBER VERSION DATE FORMAT SIZE
EVDK Based Object Counting Demonstration User Guide
FPGA-UG-02050 1.1 9/25/2018 PDF 1.3 MB
TITLE NUMBER VERSION DATE FORMAT SIZE
EVDK Based Object Counting Demonstration Bitstreams
1.1 9/25/2018 ZIP 9.7 MB


Like most websites, we use cookies and similar technologies to enhance your user experience. We also allow third parties to place cookies on our website. By continuing to use this website you consent to the use of cookies as described in our Cookie Policy.