High-Performance Distributed Multimedia Computing

1
High-Performance Distributed Multimedia Computing

• Frank Seinstra, Jan-Mark Geusebroek

MultimediaN (BSIK Project)
Intelligent Systems Lab AmsterdamInformatics
InstituteUniversity of Amsterdam
2
MultimediaN and DAS-3
3
MultimediaN and high-performance computing
Van Essen et al. Science 255, 1999.
4
A Real Problem, part 1

• News Broadcast - September 21, 2005 (see
  video1.wmv)

• Police investigating over 80.000 (!) CCTV
  recordings
• First match found no earlier than 2.5 months
  after July 7 attacks

5
Image/Video Content Analysis

• Lots of research benchmark evaluations
• PASCAL-VOC (10,000 images), TRECVID (200 hours
  of video)

• A Problem of scale
• At least 30-50 hours of processing time per hour
  of video!
• BeeldGeluid gt 20.000 hours of TV broadcasts per
  year
• NASA gt over 850 Gb of hyper-spectral image data
  per day
• London Underground gt over 120.000 years of
  processing !!!

6
High Performance Computing

• Solution
• Very, very large scale parallel and distributed
  computing
• New Problem
• Very, very complicated software

Solution tool to make parallel distributed
computing transparent to user
User
Wide-Area Grid Systems
7
Parallel-Horus Features (1)

• Sequential programming

Parallel-Horus
Sequential API
Parallelizable Patterns
Seinstra et al., Parallel Computing,
28(7-8)967-993, August 2002
8
Parallel-Horus Features (2)

• Lazy Parallelization

Seinstra et al., IEEE Trans. Par. Dist. Syst.,
15(10)865-877, October 2004
9
Extensions for Distributed Computing

• Wide-Area Multimedia Services

Parallel Horus Client
Parallel Horus Server
Parallel Horus Servers
Parallel Horus Servers
Parallel Horus Client

• User transparency?
• Abstractions techniques?
• Grid connectivity problems?

10
A Real Problem, part 2
LambdaRAM ??
may be time-critical!
11
Color-Based Object Recognition (1)

• Our Solution
• Place retina over input image
• Each of 37 retinal areas serves as a receptive
  field
• For each receptive field
• Obtain set of local histograms, invariant to
  shading / lighting
• Estimate Weibull parameters ß and ? for each
  histogram
• Hence scene description by set of 37x4x3 444
  parameters

Geusebroek, British Machine Vision Conference,
2006.
12
Color-Based Object Recognition (2)

• Learning phase
• Set of 444 parameters is stored in database
• So learning from 1 example, under single visual
  setting

a hedgehog

• Recognition phase
• Validation by showing objects under at least 50
  different conditions
• Lighting direction
• Lighting color
• Viewing position

13
Amsterdam Library of Object Images (ALOI)

• In laboratory setting
• 300 objects correctly recognized under all (!)
  visual conditions
• 700 remaining objects missed under extreme
  conditions only

Geusebroek et al., Int. J. Comput. Vis..
61(1)103-112, January 2005
14
Example Object Recognition
See also http//www.science.uva.nl/fjseins/aibo.
html
15
Example Object Recognition
(see video2.wmv)
Demonstrated live (a.o.) at ECCV 2006, June 8-11,
2006, Graz, Austria
16
Performance / Speedup on DAS-2
Single cluster, client side speedup
Four clusters, client side speedup

• Recognition on single machine /- 30 seconds
• Using multiple clusters up to 10 frames per
  second
• Insightful even distant clusters can be used
  effectively for close to real-time recognition

17
Results applicability

• Beneficial
• Performance gains largely obtained for free
• With Parallel-Horus we can build similar complex
  Grid applications in a matter of hours

18
Current Future Work

• Very Large-Scale Distributed Multimedia
  Computing
• Overcome practical annoyances
• Software portability, firewall circumvention,
  authentication,
• Optimization and efficiency
• Tolerant to dynamic Grid circumstances,
• Systematic integration of MM-domain-specific
  knowledge,
• Deal with non-trivial communication patterns
• Heavy intra- inter-cluster communication,
• Reach the end users
• Programming models, execution scenarios,

• Collaboration with VU (Prof. Henri Bal) GridLab
• Ibis www.cs.vu.nl/ibis/
• Grid Application Toolkit www.gridlab.org

19
Conclusions

• Effective integration of results from two largely
  distinct research fields
• Ease of programming gt quick solutions
• With DAS-3 / StarPlane we can start to take on
  much more complicated problems

• But most of all
• DAS-3 very significant for future MM research

20
The End
(see video3.avi)