Taming your media chaos

Preview:

DESCRIPTION

 

Citation preview

taming your media chaos

iMinds13th Oct 2011

Karel BraeckmanVRT-medialab

try finding the video about an earthquake

A computer sees bits

? earthquake

10011001010010110101010101010110101101010101010101111001

10011001010010110101010101010110101101010101010101111001

10011001010010110101010101010110101101010101010101111001

Textual metadata is needed to help the system find the correct videos

earthquake

HAITI

VICTIMS

EARTHQUAKE

how to obtain metadata?

manual

high qualitybut

takes time

9

SCAIEMEDIAMAP

PISAMEDIALOE

P

CHAMP

manual be smart

repurpose

crowd sourcing

extraction tools

linked data

manual be smart

repurpose

crowd sourcing

extraction tools

linked data

manual be smart

repurpose

crowd sourcing

extraction tools

linked data

save & structure informationwhile creating

“cradle to the grave”

be smart

be smart @ Champ

@ Champbe smart

@ MediaMapbe smart

@ PISA “Scoop”be smart

http://www.youtube.com/watch?v=cc_xQN7iyZo

= future

but requires change of toolswhat about productions

not made with those tools?

be smart

manual be smart

repurpose

crowd sourcing

extraction tools

linked data

repurpose information from existing systems for search & navigation

repurpose

@ PISA “Trouvaille”repurpose

@ MediaLoep

manual

annotations

subtitles

playout schedule

press photos

news preparation

news cms

news agencies

scripts

repurpose

lots of metadatabut

not applicable to all material

repurpose

manual be smart

repurpose

crowd sourcing

extraction tools

linked data

while interacting with your contentpeople create metadata

crowd sourcing

crowd sourcing @ DeKoers

@ DeKoers

#gilbert

#sprint

#muur

#boonen

crowd sourcing

manual be smart

repurpose

crowd sourcing

extraction tools

linked data

Speech Jingle

“Live”,”Brussel”,“Goedele Devroy”

“Ik sta in Brussel te wachten maar…”

face = “Goedele Devroy”

extraction tools

Detect shot / scene / audio changes

extraction tools @ PISA

Detect and recognize faces

extraction tools @ PISA

Detect reuse

extraction tools @ PISA

@ PISA

promising

low-level metadataacademic

extraction tools

@ EBU SCAIE

list & evaluatefeature extraction tools

academic → real-life

extraction tools

manual be smart

repurpose

crowd sourcing

extraction tools

linked data

manual be smart

repurpose

crowd sourcing

extraction tools

linked data

Enhancing existing metadata using Linked Open Data

HAITI

Plotting a location on a map requires the latitude / longitude

BAN KI-MOON

Showing a photograph or summarizing a set of search results

Open online knowledge basescontain an abundance of knowledge

DBpedia

Geo-names

W3CWordNet

FlickrWrappr

H A I T I I S A C O U N T R Y

9 M I L L I O N P E O P L E L I V E I N H A I T I

H A I T I H A S C O O R D I N AT E S 1 9 ° N 7 2 ° W

E A R T H Q A K E I S A G E O L O G I C A L P H E N O M E N O N

T H I S I S A P I C T U R ER E L AT E D T O E A R T H Q U A K E

HAITI countrytype

population8 924 000

HAITIlatitude, longitude

19° N 72° W GeoNames

sameAs

HAITI Link to information from DBpedia & GeoNamesKEYWORDS

@ MediaLoeplinked data

VISIT OUR BOOTH FOR A DEMO!

VISIT OUR BOOTH FOR A DEMO!

from PoC to product @

ALSO VISITTHIS BOOTH!

manual be smart

repurpose

crowd sourcing

extraction tools

linked data

Obtain & structure metadata: many methods, none is perfect

thank you

Recommended