ACCESSING LARGE
AUDIOVISUAL COLLECTIONS
USING VISUAL ANALYSIS
AV IN DH WORKSHOP @ DH2014 LAUSANNE
ROELAND ORDELMAN
NETHERLANDS INSTITUTE FOR
SOUND AND VISION
BUSINESS ARCHIVE
DUTCH PUBLIC BROADCASTERS
LARGE DIGITIZATION PROGRAMS
CLARIAH PRESENTATIE 11 September 2013 6
+800.000 hours of audiovisual content
‘POTENTIAL’
Find what you were (not)
looking for
Browse video to find what you
were looking for
X
We need labels!
Labels connect
(content, context)
Labeling
CLARIAH PRESENTATIE 11 September 2013 13
BIG DATA!
CLARIAH PRESENTATIE 11 September 2013 14
INNOVATIVE PLATFORMS
We need USEFUL labels
16
USEFUL?
Developer/
ICT researcher
DH Researcher
17
FEEDBACK
Research & Education
Broadcast
Professionals
Hergebruik
Media Archivists
(documentalisten)
Beschrijven
Journalists Researc...
19
Use Scenarios
&
System
Requirements
Interview
&
elicitation
sessions
Mock-up
creation &
evaluation
Prototype
evaluation...
BUILDING PROTOTYPES
2012
2013
-PRO
-RES
Onderzoekers
Media Professionals
<nisv@axes> ls –l
Total 10
-r--r--r--. 1 nisv axes 301 Jun 26 2011 METADATA
-r--r--r--. 1 nisv axes 301 Jun 26 2011 SUBTIT...
Face Recognition
Query by example
DETECTION REQUIRES TRAINING
(EXAMPLES)
2nd EC review meeting – Hilversum – Mar
19th 2013
2nd EC review meeting –
Hilversum – Mar 19th 2013
EXPECTATION MANAGEMENT
2nd EC review meeting –
Hilversum – Mar 19th 2013
Expectation Management
• Expectation management:
– Training examples versus result list
– Google images search versus visu...
DH perspective
• First explorations in various projects
– Requirements studies
– Demonstrations
– Prototypes
• Technology ...
Technology exists that could help
Technology does not solve all
problems
Discuss with ICT experts
Technology has a price, ...
www.axes-project.eu
roelandordelman.nl
Questions?
<nisv@axes>
<nisv@axes>
Accessing Large AV Collections using Visual Analysis in Digital Humanities
Accessing Large AV Collections using Visual Analysis in Digital Humanities
Accessing Large AV Collections using Visual Analysis in Digital Humanities
Upcoming SlideShare
Loading in …5
×

Accessing Large AV Collections using Visual Analysis in Digital Humanities

278 views

Published on

On the use of visual analysis technology to search audiovisual collections for research in the digital humanities. The presentation explains the audiovisual archive approach wrt access in general using visual analysis and discusses how this could fit into the practice of DH research on the basis of the results of the FP7 project AXES.

Published in: Technology, News & Politics
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
278
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
3
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Accessing Large AV Collections using Visual Analysis in Digital Humanities

  1. 1. ACCESSING LARGE AUDIOVISUAL COLLECTIONS USING VISUAL ANALYSIS AV IN DH WORKSHOP @ DH2014 LAUSANNE ROELAND ORDELMAN
  2. 2. NETHERLANDS INSTITUTE FOR SOUND AND VISION
  3. 3. BUSINESS ARCHIVE DUTCH PUBLIC BROADCASTERS
  4. 4. LARGE DIGITIZATION PROGRAMS
  5. 5. CLARIAH PRESENTATIE 11 September 2013 6 +800.000 hours of audiovisual content ‘POTENTIAL’
  6. 6. Find what you were (not) looking for
  7. 7. Browse video to find what you were looking for X
  8. 8. We need labels!
  9. 9. Labels connect (content, context)
  10. 10. Labeling
  11. 11. CLARIAH PRESENTATIE 11 September 2013 13 BIG DATA!
  12. 12. CLARIAH PRESENTATIE 11 September 2013 14 INNOVATIVE PLATFORMS
  13. 13. We need USEFUL labels
  14. 14. 16 USEFUL? Developer/ ICT researcher DH Researcher
  15. 15. 17 FEEDBACK
  16. 16. Research & Education Broadcast Professionals Hergebruik Media Archivists (documentalisten) Beschrijven Journalists Research Academic researchers Investigate Education Illustrate
  17. 17. 19 Use Scenarios & System Requirements Interview & elicitation sessions Mock-up creation & evaluation Prototype evaluation System evaluation Surveys & log analysis Qualitative Qualitative QualitativeQuantitative Quantitative/ Qualitative
  18. 18. BUILDING PROTOTYPES
  19. 19. 2012 2013 -PRO -RES Onderzoekers Media Professionals
  20. 20. <nisv@axes> ls –l Total 10 -r--r--r--. 1 nisv axes 301 Jun 26 2011 METADATA -r--r--r--. 1 nisv axes 301 Jun 26 2011 SUBTITLES -r--r--r--. 1 nisv axes 301 Jun 26 2011 SPEECH RECOGNITION -r--r--r--. 1 nisv axes 301 Jun 26 2011 FACE RECOGNITION -r--r--r--. 1 nisv axes 301 Jun 26 2011 VISUAL CONCEPT DETECT -r--r--r--. 1 nisv axes 301 Jun 26 2011 EVENT DETECTION -r--r--r--. 1 nisv axes 301 Jun 26 2011 LOCATION DETECTION -r--r--r--. 1 nisv axes 301 Jun 26 2011 QUERY BY EXAMPLE -r--r--r--. 1 nisv axes 301 Jun 26 2011 SEARCH -r--r--r--. 1 nisv axes 301 Jun 26 2011 RECOMMENDATION -r--r--r--. 1 nisv axes 301 Jun 26 2011 USER INTERFACE <nisv@axes> |
  21. 21. Face Recognition
  22. 22. Query by example
  23. 23. DETECTION REQUIRES TRAINING (EXAMPLES) 2nd EC review meeting – Hilversum – Mar 19th 2013
  24. 24. 2nd EC review meeting – Hilversum – Mar 19th 2013 EXPECTATION MANAGEMENT
  25. 25. 2nd EC review meeting – Hilversum – Mar 19th 2013
  26. 26. Expectation Management • Expectation management: – Training examples versus result list – Google images search versus visual search in AV • Understanding visual search: – why something is hard to detect • visual characteristics, training examples – Noise is not bad per definition
  27. 27. DH perspective • First explorations in various projects – Requirements studies – Demonstrations – Prototypes • Technology is ready to start exploring its use in real use scenarios (e.g., query by example) • Feed DH ideas into ICT research community
  28. 28. Technology exists that could help Technology does not solve all problems Discuss with ICT experts Technology has a price, what is the RoI? AWARENESS How does technology fit in How do limitations fit in ‘Technology Critique’ (Historian 2.0?) ICT and curriculum METHODOLOGY/TRAINING What can it do? How does it work? How does it perform? How can it be improved? MICRO MACRO How can we use it? What do we need? How does it scale? Who could benefit as well?
  29. 29. www.axes-project.eu roelandordelman.nl Questions? <nisv@axes> <nisv@axes>

×