Pervasive Artificial Intelligence Research (PAIR) Labs

Drones and Computer Vision

Pervasive Artificial Intelligence Research (PAIR) Labs

Drones and Computer Vision

NCTU – Professor Jen-Hui Chuang

Principal Investigator: Jen-Hui Chuang

Research :

Development of AI platform of Smart Drone ─ Intelligent Fight Control

Core Technologies

The techniques we developed include:
1. the camera calibration based on geometry principle
2. the one-shot fisheye camera calibration system
3. the multi-intensity with long-distance illuminated IR surveillance system and objects detection method
4. the smartglasses-based egocentric ranging/pointing system
5. the vision-based antenna pointing and registration system
6. multi-camera pedestrian positioning system

Applications

1. Camera calibration may apply to automatic optical inspection(AOI) and vehicle camera system.
2. Smartglasses-based pointing system may apply to guiding system
3. ulti-intensity with long-distance illuminated IR surveillance system may apply to indoor/outdoor surveillance system.
4. Vision-based antenna pointing and registration system may apply to high-speed wireless transmission on UAV.
5. Multi-camera pedestrian positioning system may apply to indoor surveillance and epidemic prevention.

Technology Transfer

1. Fast and high-precision camera calibration system
2. One-shot fisheye camera calibration system
3. Multi-intensity with long-distance illuminated IR surveillance system
4. Smartglasses-based egocentric ranging/pointing system
5. Vision-based antenna pointing and registration system
6. Multi-camera pedestrian positioning system
7. multi-camera pedestrian positioning system

Technologies with Potential for Technology Transfer or Industry-Academia Collaboration

1. Fully automatic camera calibration system by using virtual patterns on display.
2. Fisheye stereo without rectification
3. High-performanced face/human detection in IR surveillance system
4. Multiple planes/surface pointing system.
5. Simple camera-eye (epipolar) geometry calibration algorithm.
6. Multi-machine interconnection pointing system.
7. Multiple antenna/base station antenna pointing system.
8. Multi-camera pedestrian positioning system using deep learning to enhance efficiency and accuracy

Proposal for Industry-Academia Collaboration

1. Use fast and high-precision camera calibration system in AOI system
2. Use the one-shot fisheye camera calibration system to production line of 360-camera.
3. Use multi-intensity with long-distance illuminated IR surveillance system to detect suspicious characters at night time.
4. Use smartglasses-based pointing system to develop human-computer interface, e.g. AR wearable devices, operation instruction system on production line, shooting games, etc.
5. Use vision-based antenna pointing and registration system to apply to high-speed wireless transmission on UAV.