All Projects → devolksbank → Raspberry-Pi-Movidius-Person-Detector

devolksbank / Raspberry-Pi-Movidius-Person-Detector

Licence: MIT License
Raspberry Pi Movidius Neural Compute Stick person detector installer

Programming Languages

shell
77523 projects

Projects that are alternatives of or similar to Raspberry-Pi-Movidius-Person-Detector

book-ml
書籍「今すぐ試したい!機械学習・深層学習(ディープラーニング)画像認識プログラミングレシピ」のソースコードを配布するレポジトリです。
Stars: ✭ 29 (+45%)
Mutual labels:  raspberry-pi-3, movidius
raspiApWlanScripts
Wifi station/AP mode scripts
Stars: ✭ 24 (+20%)
Mutual labels:  raspberry-pi-3
deeper-traffic-lights
[repo not maintained] Check out https://diffgram.com if you want to build a visual intelligence
Stars: ✭ 91 (+355%)
Mutual labels:  objectdetection
Face-Recognition-System
Intelligent Facial Recognition with Django Restful IoT on Raspberry Pi
Stars: ✭ 46 (+130%)
Mutual labels:  movidius
LetsHack
Notes & HowTo's covering the Raspberry Pi, Arduino, ESP8266, ESP32, etc.
Stars: ✭ 37 (+85%)
Mutual labels:  raspberry-pi-3
PiBuilder
Ideas for building a Raspberry Pi from "bare metal" to ready-to-run IOTstack
Stars: ✭ 26 (+30%)
Mutual labels:  raspberry-pi-3
gro-light-automation
A raspberry pi project to automate hydroponics with relays and data sensors through a web application
Stars: ✭ 44 (+120%)
Mutual labels:  raspberry-pi-3
homeberry
HomeBerry is an Android remote control app for your Raspberry PI
Stars: ✭ 31 (+55%)
Mutual labels:  raspberry-pi-3
ros openvino
A ROS package to wrap openvino inference engine and get it working with Myriad and GPU
Stars: ✭ 57 (+185%)
Mutual labels:  movidius
raspberry-pi-3-setup
My Raspberry Pi 3 setup instructions and notes
Stars: ✭ 35 (+75%)
Mutual labels:  raspberry-pi-3
pedalevite
Pédale Vite — DIY multi-FX pedalboard for guitar/bass/etc.
Stars: ✭ 68 (+240%)
Mutual labels:  raspberry-pi-3
mmpm
MagicMirror Package Manager
Stars: ✭ 104 (+420%)
Mutual labels:  raspberry-pi-3
vigilantpi
VigilantPI is a lightweight NVR which targets mainly Raspberry Pi
Stars: ✭ 14 (-30%)
Mutual labels:  raspberry-pi-3
RaspberryPi-Packet-Sniffer
An HTTP and HTTPS sniffing tool created using a Raspberry Pi
Stars: ✭ 79 (+295%)
Mutual labels:  raspberry-pi-3
Human-detection-system-with-raspberry-Pi
A motion detection system with RaspberryPi, OpenCV, Python
Stars: ✭ 50 (+150%)
Mutual labels:  raspberry-pi-3
SwiftySHT20
⚡ A Swift library for the I2C SHT20 Humidity and Temperature Sensor.
Stars: ✭ 19 (-5%)
Mutual labels:  raspberry-pi-3
gitlab-runner
GitLab Runner (Docker image) for ARM devices, this is a mirror repository of
Stars: ✭ 17 (-15%)
Mutual labels:  raspberry-pi-3
Lumos
😴 Smart Lighting for Better Sleep
Stars: ✭ 98 (+390%)
Mutual labels:  raspberry-pi-3
imusensor
Python library for communication between raspberry pi and MPU9250 imu
Stars: ✭ 47 (+135%)
Mutual labels:  raspberry-pi-3
SwiftFlowMeter
⚡️ A Swift library for using Hall effect based water flow sensors.
Stars: ✭ 22 (+10%)
Mutual labels:  raspberry-pi-3

Intel Movidius Neural Compute Stick Raspberry-Pi Webcam Installer

This project allows you to install a fully working person-detector using a single installation script.

We created this script so anyone can easily try this software/hardware out and experience what AI can do, to help you imagine what else you can do with it to help you in whatever project you're interested in!

More information in our blog (NL)

raspberry-pi compute-stick webcam

Prerequisites

For this project you need (at least):

And preferrably also:

  • a separate device (laptop/desktop) to access the cam stream remotely

Installation steps

For the installation, you need to perform the following steps

1) Start strech OS on the Raspberry-Pi

  • download strech OS
  • burn the iso onto the SD card, for example using Etcher
  • put the SD card into the Raspberry-Pi
  • connect the keyboard and mouse to the Raspberry-Pi
  • use the hdmi cable to connect the monitor and the Raspberry-Pi
  • use the power adapter to connect the Raspberry-Pi to a power outlet
  • wait until it has started
  • go to the preferences menu option on the Raspberry-Pi and start 'Raspberry Pi Configuration'
  • go to the localisation tab en set your wifi country
  • connect the Raspberry-Pi to your wifi using the icon on the top of your screen
  • connect the compute stick and the webcam to the Raspberry-Pi

Enable ssh access

The easiest way to control the installation is from your PC/laptop, so it is advisable to enable SSH access on the pi, but you can also run everything on the Raspberry-Pi itself:

  • go to the preferences menu option on the Raspberry-Pi and start 'Raspberry Pi Configuration'
  • on the interfaces tab, enable 'ssh'
  • check your IP address using ifconfig in a terminal
  • on your PC/laptop, connect using ssh pi@<ip> using the default password raspberry (to be changed as you wish)

2) Update the firmware

Run in a terminal:

sudo rpi-update

And reboot afterwards:

sudo reboot

3) Clone this repository

git clone https://github.com/devolksbank/Raspberry-Pi-Movidius-Person-Detector

4) Call the installation script

The installation script can be called step-by-step, using:

chmod u+x install*
./install-step1-ncs-api.sh
./install-step2-image-classifier.sh
./install-step3-opencv.sh
./install-step4-webcam.sh

Or by the single-command installer:

chmod u+x install*
./install-full.sh

5) Test the installed streamer

First source the ~/.profile to get the references to opencv right. Next, call:

cd ~/workspace/ncappzoo/caffe/NCS-Pi-Stream/ && python3 streamer_ncs.py

And watch the output via http://localhost:8080/cam.mjpg or remote via http://<ip>:8080/cam.mjpg. Note that when watching via localhost, your Raspberry-Pi can quickly overheat!

Thanks

Many thanks to the creators of these blogs/articles/repositories, which made creating this repository a lot easier:

Contact

For any questions please don't hesitate to contact us at [email protected]

logo

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].