Seminar on Media Technology Computer Vision Albert Alemany Font
Outlines ❖ Introduction What is computer vision and why this topic ❖ History of computer vision and related disciplines ❖ Applications Face/smile detection, OCR, object recognition, medical imaging,... ❖ Conclusions ❖ References
What is computer vision? ❖ Traffic scene ❖ Number of vehicles ❖ Type of vehicles ❖ Location of closest obstacle ❖ Assessment of congestion ❖ Location of the scene captures ❖... Given an image or more, extract properties of the 3D world
Why this topic?
Related disciplines
History of computer vision ❖ 1950′s – Two dimensional imaging for statistical pattern recognition developed ❖ 1960′s – Roberts begins studying 3D machine vision ❖ 1970′s – MIT’s Artificial Intelligence Lab opens a "Computer Vision" course ❖ 1980’s – New theories and concepts emerging. Shift toward geometry and increased mathematical rigor ❖ 1990’s – Face recognition. Statistical analysis in vogue ❖ 2000’s – Broader recognition. Large annotated datasets available. Video processing starts
Finding people in images "Yes" instances
Finding people in images "No" instances
Face detection ❖ The camera detects faces in a scene and then automatically focus (AF) and optimizes exposure (AE) and, if needed, flash output Face detection in digital cameras
Smile detection
Optical character recognition (OCR) Technology to convert scanned docs to text
Vision-based biometrics Photographer: Steve McCurry How the Afghan girl was identified by her iris pattern: Right eye processed image Right eye processed image
Object recognition ❖ Google goggles Query image Webpage Matching image ❖ Lincoln Microsoft Research
Mimic human behaviour?
Limits of human vision
Vision evolution Google reCaptcha
Making the invisible visible Eulerian Video Magnification for Revealing Subtle Changes in the World SIGGRAPH Raw version
Making the invisible visible Eulerian Video Magnification for Revealing Subtle Changes in the World Magnified version SIGGRAPH 2012
Smart cars
Medical imaging Image guided surgery3D Imaging
Special effects: shape capture The Matrix movies, ESC Entertainment
Special effects: shape capture
Special effects: motion capture Pirates of the caribbean, Industrial Light and Magic
Video-based interaction: gaming Sony Eyetoy Microsoft Natal
Image mosaic ❖ 3D from multiple images ❖ 3D from one image ❖ "Big" image from other images/video
Image mosaic
Supermarket scanner
Conclusions
References ❖ Richard Szeliski (2010). Computer Vision: Algorithms and Applications. Springer-Verlag. ❖ Gérard Medioni and Sing Bing Kang (2004). Emerging Topics in Computer Vision. Prentice Hall. ❖ Pedram Azad, Tilo Gockel, Rüdiger Dillmann (2008). Computer Vision – Principles and Practice. Elektor International Media BV. ❖ ❖
Thank you for your attention