IEEE paper web image re-ranking using query specific semantic signatures.. presentation for MCA project. SIFT algorithm- scale invariant feature transform.

1. Dept. of CSE, MCET PTA 1
LEKSHMI R S
Register No: 400901
IMAGE HUB
(WEB IMAGE RERANKING USING
QUERY SPECIFIC SEMANTIC
SIGNATURES)

2. 2
o Effective way to improve the results of web-based image
search
o Automatically offline learns different visual semantic
spaces for different query keywords
o Through keyword expansions
o Improves both the accuracy and efficiency of image re-
ranking.
IMAGE HUB
Dept. of CSE, MCET PTA

3.  Query keyword- keyword expansion.
 Reference classes.
 Removing outliers.
 Top images as training examples.
 Multiclass classifiers of low level visual
features.
 Semantic signature is extracted.
 Images re-ranked based on semantic signatures.
3
IMAGE HUB
Dept. of CSE, MCET PTA

4. NO
SIMILARITY
SCORE OUTLIERS
4
IMAGE HUB
Dept. of CSE, MCET PTA

5. 5
OUTLIERS 80%-90%
IMAGE HUB
Dept. of CSE, MCET PTA

6. 1. Image Search
2. Images Retrieved by
Expanded Keywords
3. Query Categorization
4. Visual Query Expansion
6
IMAGE HUB
Dept. of CSE, MCET PTA

7. 7
SIFT
SCALE INVARIANT FEATURE TRANSFORM
IMAGE HUB
Dept. of CSE, MCET PTA

8. SCALE INVARIANT
FEATURE
DETECTION
FEATURE
MATCHING AND
INDEXING
CLUSTER
IDENTIFICATION
MODEL
VERIFICATION
OUTLIER
DETECTION
8
IMAGE HUB
Dept. of CSE, MCET PTA

9.  Software Specification :
 Front End : Java
 Back End : My SQL
 Operating System : Windows/Linux
 IDE : NetBeans
 Hardware Specification :
 Processor : Pentium IV OR Above
 Primary Memory : 1GB RAM
 Storage : 40 GB Hard Disk
 Display : VGA ColorMonitor
 Key Board : Windows compatible
 Mouse : Windows compatible
9
IMAGE HUB
Dept. of CSE, MCET PTA

10. 10
USECASE DIAGRAM
IMAGE HUB
Dept. of CSE, MCET PTA

11. 11
IMAGE HUB
Dept. of CSE, MCET PTA

12. 12
DFD LEVEL 0
IMAGE HUB
Dept. of CSE, MCET PTA

13. 13
DFD LEVEL 1- ADMIN
IMAGE HUB
Dept. of CSE, MCET PTA

14. 14
DFD LEVEL 2- ADMIN
IMAGE HUB
Dept. of CSE, MCET PTA

15. 15
DFD LEVEL 2- ADMIN
IMAGE HUB
Dept. of CSE, MCET PTA

16. 16
DFD LEVEL 1- USER
IMAGE HUB
Dept. of CSE, MCET PTA

17. 17
DFD LEVEL 2- USER
IMAGE HUB
Dept. of CSE, MCET PTA

18. 18
DFD LEVEL 2- USER
IMAGE HUB
Dept. of CSE, MCET PTA

19. 19
IMAGE HUB
Dept. of CSE, MCET PTA

20. • Images re-ranked by comparing similarity
scores.
• Pre-computed semantic signatures.
• Faster and efficient.
• 80%- 90% outliers removed.
20
IMAGE HUB
Dept. of CSE, MCET PTA

21. 21
IMAGE HUB
Dept. of CSE, MCET PTA
SCREENSHOTS

22. 22
HOME PAGE
IMAGE HUB
Dept. of CSE, MCET PTA

23. 23
REGISTRATION FORM
IMAGE HUB
Dept. of CSE, MCET PTA

24. 24
LOGIN FORM
IMAGE HUB
Dept. of CSE, MCET PTA

25. 25
ADMIN- ADD IMAGE
IMAGE HUB
Dept. of CSE, MCET PTA

26. 26
ADMIN- VIEW USERS
IMAGE HUB
Dept. of CSE, MCET PTA

27. 27
USER- IMAGE SEARCH
IMAGE HUB
Dept. of CSE, MCET PTA

28. 28
USER- IMAGE RERANKING
IMAGE HUB
Dept. of CSE, MCET PTA

29. 29
USER- ADVANCED WEB SEARCH
IMAGE HUB
Dept. of CSE, MCET PTA

30. 1. Web image re-ranking using query specific semantic signatures,
IEEE Transactions on Pattern Analysis and Machine Intelligence
(Volume:36 , Issue: 4 ) April 2014, DOI:10.1109/TPAMI.201.
2. E. Bart and S. Ullman. Single-example learning of novel classes
using representation by similarity. In Proc. BMVC, 2005.
3. Y. Cao, C. Wang, Z. Li, L. Zhang, and L. Zhang. Spatial-bag-of
features. In Proc. CVPR, 2010.
4. G. Cauwenberghs and T. Poggio. Incremental and decremental
support vector machine learning. In Proc. NIPS, 2001.
30
Dept. of CSE, MCET PTA
IMAGE HUB

31. 31
IMAGE HUB
Dept. of CSE, MCET PTA