Introduction

Based on computer vision algorithms, this software enables to synchronize slides with the corresponding video recording.

Note

slideotracker = slides + video + tracking

• Author : Sebastien Campion / INRIA Texmex Team / http://www.irisa.fr/texmex

Licence

GNU AFFERO GENERAL PUBLIC LICENSE v3

Screencast

Download

Debian package

make deb

Python packages

make pydist

Source code

https://github.com/scampion/slideotracker

Usage

Usage: slideotracker.py [options]

Options: -h, --help show this help message and exit -i index, --index=index

index file is a simple text file, the first line in the video path, others lines are paths on slide images (use ImageMagick, to convert pdf in several images)

-p PRECISION, --precision=PRECISION
	precision in number of frame (default 25)
-o OUTFILE, --out=OUTFILE
	output file name, by default results.js
-f FORMAT, --format=FORMAT
	output file format js (default),csv
-d, --debug	debug trace

Usage

Usage: slideotracker.py [options]

Options: -h, --help show this help message and exit -i index, --index=index

index file is a simple text file, the first line in the video path, others lines are paths on slide images (use ImageMagick, to convert pdf in several images)

-p PRECISION, --precision=PRECISION
	precision in number of frame (default 25)
-o OUTFILE, --out=OUTFILE
	output file name, by default results.js
-f FORMAT, --format=FORMAT
	output file format js (default),csv
-d, --debug	debug trace

Example

Real test

Download and extract :

http://slideotracker.gforge.inria.fr/demo/demo.tar.gz

Run

slideo -i tests/data/test.txt

Common use

#extract pdf pages to jpeg using ImageMagick
mkdir /tmp/mypdf/
convert <your.pdf> /tmp/mypdf/slide.jpg
#build a index file
echo './path/to/my/video.ogv' > /tmp/index
ls -1 /tmp/mypdf/slide* >>/tmp/index
#Run the tracker
python slideotracker.py -i /tmp/index -o results.txt

try also :

python slideotracker.py -i tests/data/short_test.txt

Results in Javascript format :

slides=[0, 1, 2];

frames=[500, 850, 950];

Results in CSV format :

#slide_number;star_frame;end_frame

0;0;500

1;500;850

2;850;950

... .. .

Display results in HTML5

Open the following file with your browser

/usr/share/slideo/html/player.html?mediafile=data/video.ogv&slidedir=data/slides&fps=25

Dependencies

• scikit-learn
• OpenCV 2.1

TODO/Roadmap

• geometric robustification ... in progress
• optimize time computing
• documentation / How it works

Credits

• Images used for the logo : Dropline Nuovo! from http://art.gnome.org/themes/icon
• RMLL 2010 video for the data test