Skip to content

Latest commit

 

History

History
85 lines (65 loc) · 3.5 KB

README.md

File metadata and controls

85 lines (65 loc) · 3.5 KB
Raw

PHODO : Face Tracking System

Contributors & Oragnization

Contributors

  • Keunju Song
  • Yechan Yun
  • HyeKyung Kim
  • SoHee Park
  • JungYoong Min
  • Su Min Kim (adviser)

Oragnization

  • Tech University of Korea LINC+
  • 20th Industrial Technology Exhibition, Tech University of Korea

Overview

PHODO is face tracking system that consisted of the main body and mobile app.

PHODO's face tracking technology is based on Machine Learning.

From this system, we can anticipate applications in Entertainment and Security field with facial similarity.

Paper about PHODO is updated. [Click]

See more information about PHODO : https://yyc9920.github.io/phodo.github.io/index.html

System Architecture

System Block Diagram

H/W

  • Raspberry Pi Zero W (MCU)
  • Stepping Motors (2EA)
  • Servo Motor
  • LiPo battery
  • Rechargeable 5V Lipo USB Boost - Adafruit
  • Ultrasonic Sensor - HC-SR04
  • Tripod
  • Camera Rail

Application

PHODO

  • Supporting basic camera function
  • Tracking Faces
  • Supporting Buletooth that links to the main body

3D Modeling

  • Designed by JungYoong Min

Function of the System

  • Multiple Face Tracking: When there are more than one face on the screen, tracking the coordinates of the faces by averaging operations.
  • User's Face Customizing: PHODO can recognize the face individually, so it can tracking just one person that user want.
  • Handsfree Capture: PHODO have voice recognition to support the handsfree capture. Setting voice language is Korean.
  • 3-Div Optimal Composition: PHODO can take the photo automatically with optimal composition by controling face coordinates. Face position is established Left-Center-Right.

Result

  • PHODO main body

  • PHODO Application

  • PHODO Face Tracking action

Development Environment

  • Android Studio IDE (Java, C++, Kotlin)
  • Raspberry Pi Terminal (Python)

Reference