Upload
others
View
6
Download
0
Embed Size (px)
Citation preview
Image Processing LabDipartimento di Matematica e Informatica
Università degli Studi di Catania
Antonino Furnari
10/05/2014
An Introduction to Augmented Reality
Computer Vision AA. 2013/2014 - Augmented Reality
Augmented Reality
“a live, copy, view of a physical, real-world environment whose elements are augmented (or supplemented) by computer-generated sensory
input such as sound, video, graphics or GPS data”
“it is related to a more general concept called mediated reality, in which a view of reality is
modified (possibly even diminished rather than augmented) by a computer”
“as a result, the technology functions by enhancing one’s current perception of reality.”
wikipedia definition
Computer Vision AA. 2013/2014 - Augmented Reality
Applications
Computer Vision AA. 2013/2014 - Augmented Reality
Applications
Computer Vision AA. 2013/2014 - Augmented Reality
Some History
● The term “augmented reality” appears since the 1940s;
● The first augmented head mounted display is invented by Ivan Sutherland in 1968;
● First systems using mobile devices, internet and geolocalization appear in the 90s;
● Advances in the 2000s;● Augmented Reality diffusion in the 2010s.
Computer Vision AA. 2013/2014 - Augmented Reality
Hardware
Some technologies which make AR interesting:● Head-mounted (with 6 DoF monitoring);● Handheld:
– Mobile phones;– Tablets;
● Wearable devices:
– Google glass;– Orcam (video - http://www.orcam.com/);– Epson moverio.
Computer Vision AA. 2013/2014 - Augmented Reality
Computer Vision based Augmented Reality
● Computer Vision allows to create augmented reality applications by superimposing 2D or 3D contents on the scene;
● In order to do so we need to:
1) detect and track the area where to show the content;
2) estimate its 3D position in the real world;
3) render the 2D/3D content according to the estimated position;
● Two main technologies:
– fiduciary markers;– markerless (i.e., object detection).
Computer Vision AA. 2013/2014 - Augmented Reality
Fiduciary Markers Augmented Reality: ARToolkit
● ARToolkit is an Open Source toolkit for marker-based augmented reality;
● It is quite old (last update in 2007) but still a good starting point for understanding the AR concepts (open & well documented);
● It offers functions for detecting and tracking single or multiple markers while relaying on OpenGL/glut for 2D/3D rendering;
● http://www.hitl.washington.edu/artoolkit.
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkitdemo
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolit
● For additional informations about the next topics, the reader is referred to the very well written ARToolkit documentation and tutorials:
– http://www.hitl.washington.edu/artoolkit/documentation/
● Some other useful information can be found in the examples provided with the toolkit.
documentation
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkit
● Follow the instructions at:
– http://www.hitl.washington.edu/artoolkit/documentation/usersetup.htm
● For Linux/GStreamer configuration, include the following line (check your video device) in your .bashrc file:
– export ARTOOLKIT_CONFIG="v4l2src device=/dev/video0 usefixedfps=false ! ffmpegcolorspace ! capsfilter caps=video/xrawrgb,bpp=24 ! identity name=artoolkit ! fakesink"
setup
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkitbasic principles
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkitthe marker
● The marker plays the role of an object which geometry is known;
● In particular:
– we chose markers which are easily detectable (thick black borders);
– we know the real world size of the marker;– we chose the inner symbol which is neither
horizontally nor vertically symmetric in order to estimate its rotation.
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkitmarker detection
original image thresholded image connected components
contours edges and corners fitted square
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkitmaker candidates
● This method actually allows to find just candidates: any pattern with thick black borders;
● The actual marker is found normalizing the candidates and comparing them with the searched pattern using template matching;
● The candidate giving the highest confidence is selected.
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkitmarker matching
0.3
0.92
0.2
0.4
normalization
...
...
...
searched pattern
normalized candidates
found candidates
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkit3D position and orientation estimation
● Now that we have an object which geometry, size, position and orientation are known, we can estimate its 3D position with respect to the camera;
● It can be done computing the extrinsic parameters as seen for camera calibration;
● Intrinsic parameters which are good for most cameras are part of the toolkit. Specific parameters can be obtained calibrating the camera.
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkitcoordinate systems
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkitcamera and marker relationships (demo - exview)
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkitarchitecture
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkitdevelopment principles
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkitinto the code!
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkit
● Facilities are included in the toolkit to enable multi-marker tracking;
● A multi-marker pattern is used where the relative position of the markers is known;
● The transformation matrices between the different markers must be supplied.
multiple markers (demo)
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkittrain a new pattern
● In order to use your own pattern, the blank pattern has to be modified and printed;
● The provided mk_patt application allows to generate a patt.yourpatt file containing all the information needed for the pattern matching.
Computer Vision AA. 2013/2014 - Augmented Reality
ARToolkitcamera calibration
● Intrinsic parameters which are enough general to work with most of the cameras are available in the toolkit;
● Although, in order to improve the detection and tracking performances, a utility for camera calibration is provided in order to calibrate your own camera.
Computer Vision AA. 2013/2014 - Augmented Reality
Markerless Augmented Reality?
Tracking a number of feature points (e.g., SIFT) in order to detect a marker object (e.g., a photo) and to estimate its position and orientation.
Some mobile frameworks available (e.g., Vuforia).
Computer Vision AA. 2013/2014 - Augmented Reality
Question Time
Computer Vision AA. 2013/2014 - Augmented Reality
Contacts
● For any doubts please contact me:
– Room 30;● Slides available at:
– Studium course page:● http://studium.unict.it/dokeos/2014/course
s/73072C2/