Dot Vision

An innovative interactive art installation designed for a coffee shop setting. This project aims to engage patrons by visually representing their movements as dots on a display screen. Beyond its artistic appeal, Dot Vision serves as a dynamic marketing tool for Consult NTA, showcasing the company's expertise in technology solutions.

Features

• Interactive Art: Utilizes advanced object detection and computer vision technologies to track the movements of coffee shop patrons, translating these into captivating visual representations (dots) on a digital canvas.
• Engagement Tool: Encourages interaction by allowing patrons to influence the art through their movement, creating a unique, dynamic experience.
• Educational Aspect: Includes a QR code linked to information about Dot Vision, offering insights into the technology behind the installation and highlighting Consult NTA’s capabilities.

Deployment

Deployment of this project onto a coral board consists of these steps

Get into your board

mdt shell

Run the initialization script

wget https://raw.githubusercontent.com/alvinwatner/dot-vision/main/init.sh -O - | bash 

Run Locally

Clone the project

  git clone https://github.com/alvinwatner/dot-vision

Go to the project directory

  cd dot-vision

Install dependencies

  pip install -r requirements.txt

Start the program

  python3 dot_vision.py --accelerator cpu

Options Summary

Some options (flags) can be passed into the program in order to change its behavior.

optional arguments:
  -h, --help                    show this help message and exit
  --vidsource VIDSOURCE         Video source for tracking
  --layout2Ddir LAYOUT2DDIR     2D layout image
  --layout3Ddir LAYOUT3DDIR     3D layout image
  --coor2Ddir COOR2DDIR         2D coordinates data
  --coor3Ddir COOR3DDIR         3D coordinates data
  --live                        Enable live tracking
  --modeldir MODELDIR           Directory containing the detect.tflite and labelmap.txt
  --threshold THRESHOLD         Set the threshold for object tracking accuracy
  --accelerator {cpu,tpu}       Set the accelerator used in object detection

The options can be seen by running dot_vision.py -h