Human Presence Detection

Lattice sensAI Reference Design

Customizable to detect any object: This Reference Design continuously searches for the presence of a human via a CMOS image sensor, and reports results by lighting LEDs on the board. This Reference Design is a design example that can be used as a base for your own custom solution. For an implementation of this Reference Design, see this Demo. By updating the training model using the deep learning frameworks such as Caffe or Tensorflow, your AI systems can detect and locate any object of interest.

Optimized Power and Performance - The reference design enables a flexible systems design trading off speed and power, few 1 mW for most applications.

Complete Reference Design - Training dataset, scripts for training using common Neural Network training tools and NN model are provided to enable modification.


  • Accelerated, low-power human presence detection at the network edge using neural network model
  • iCE40 UltraPlus
    • VGG8 like 16-bit CNN
    • 64*64*3 input
    • 6 zone searching
    • Up to 8 frames per second
    • 7 mW of power consumption
  • Adjustable Frame-rate. Can be optimized between power and response time depending on system needs
Lattice sensAI

This reference design is supported with 1.1 version of the Radiant Software Tool, this version of the tool can be downloaded on the Software Archive.

Jump to

Block Diagram

Human Presense Detection Reference Design Block Diagram using iCE40 UltraPlus


Quick Reference
Technical Resources
Select All
iCE40 UltraPlus Human Presence Detection Quick Start Guide
FPGA-AN-02005 2.1 10/23/2019 PDF 1.1 MB
Select All
Human Presence Detection Using Compact CNN Accelerator IP - Documentation
FPGA-RD-02059 2.1 10/23/2019 PDF 3.3 MB
Human Presence Detection Using Compact CNN Accelerator IP Detect Low Power - Project Files
2.1 10/23/2019 ZIP 16 MB

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