Seeing with Python
summary
Built a camera sensor for a Ninja Block that starts as a USB webcam (too slow) then moves to a dedicated BeagleBone Black and finally a Raspberry Pi with the Pi camera module. Uses Python bindings for OpenCV and SimpleCV plus numpy/pillow; implements motion detection, object tracking and Haar-based face detection and frame differencing. Deals with lighting and dependency issues, reaches about 30 fps on the right hardware, and ultimately works though deployment quirks remain.
AI-written summary · about
chapters (12)
- 0:00Intro
- 1:23What is Computer Vision - 1980's
- 2:58What is Computer Vision - 2014
- 3:52Computer Vision - My Reality
- 6:35CV is a lot more than grabbing an image
- 7:55Python Imaging Library
- 13:12Scipy ndimage
- 15:53SimpleCv
- 17:24OpenCV Python Wrappers
- 22:22Raspberry Pi CV
- 25:57Resources
- 26:42Contact Details
related talks
37mUnstructured Data Processing with a Raspberry Pi AI Kit and Python
40mThe eye of the Python, an eye tracking system. From zero to...
36m"Lights, camera, AI!" - Automated sports videography
31mExploring Real-time Computer Vision Using ActionCable
28mChasing birds with code
01h 12mBuilding a live face recognition system
35mDingoCar: Perception module Small scale, self-driving car
14mComputer vision on edge devices