Vision Ui

This is a vision-based 3d model manipulation and control UI

Overview

Manipulation of 3D Models Using Hand Gesture

This program allows user to manipulation 3D models (.obj format) with their hands. The project support both the OAK-D and OAK-D-LITE.

Install dependencies

On an Intel-based macOS or Linux machine, run the following command in the terminal:

git clone https://github.com/cortictechnology/vision_ui.git
cd vision_ui
python3 -m pip install -r requirements.txt

For Linux only, make sure your OAK-D device is not plugged in and then run the following:

echo 'SUBSYSTEM=="usb", ATTRS{idVendor}=="03e7", MODE="0666"' | sudo tee /etc/udev/rules.d/80-movidius.rules
sudo udevadm control --reload-rules && sudo udevadm trigger

To run

1. Make sure the OAK-D/OAK-D-Lite device is plug into the computer.
2. In the terminal, run

python3 main.py

AI Model description

The ai_models folder includes two Intel Myriad X optimized models:

1. palm_detection_sh4.blob: This is the palm detection model
2. hand_landmark_sh4.blob: This is the model to detect the hand landmarks using the palm detection model

Credits

• Google Mediapipe
• depthai_hand_tracker from geaxgx

Files in this package

• CAD source: bunny.obj, dragon.obj, elephant.obj
• Images: cursor.png

---

Source & license

• Original author: cortictechnology
• Source repository: https://github.com/cortictechnology/vision_ui
• License: MIT

Imported into the CommunityCAD Archive with attribution preserved. All rights remain with the original author under the stated license.