SLIDE 1
Outline
- Motivation
- Object Segmentation
- Experiments
- Results
- Conclusions
- Ongoing work
2
segmentation with a game Vincent Charvillat Amaia Salvador Aguilera - - PowerPoint PPT Presentation
segmentation with a game Vincent Charvillat Amaia Salvador Aguilera - - PowerPoint PPT Presentation
Crowdsourced object segmentation with a game Vincent Charvillat Amaia Salvador Aguilera Oge Marques Axel Carlier Xavier Gir i Nieto Outline Motivation Object Segmentation Experiments Results Conclusions Ongoing work
SLIDE 2
SLIDE 3
Motivation
?
3
SLIDE 4
Motivation
?
4
SLIDE 5
Semi-Supervised object segmentation
- B. C. Russell, A. Torralba, K. P. Murphy, and W. T. Freeman. Labelme:
A database and web-based tool for image annotation. IJCV, 2008
Rough segmentation
5
SLIDE 6
Semi-Supervised object segmentation
- P. Arbelaez and L. Cohen. Constrained image segmentation from hierarchical boundaries. In
CVPR'08, 2008.
- 2) K. McGuinness and N. E. O'Connor. A comparative evaluation of interactive segmentation
algorithms. 6
SLIDE 7
Semi-Supervised object segmentation
Boring task for users!
7
SLIDE 8
8
SLIDE 9
Games with a purpose
- J. Steggink and C. Snoek. Adding semantics to image-region annotations with the name-it-
- game. Multimedia Systems, 2011.
- L. von Ahn, R. Liu, and M. Blum. Peekaboom: a game for locating objects in images. In CHI'06,
2006. 9
SLIDE 10
Ask’nSeek
- A. Carlier, O. Marques, and V. Charvillat. Ask'nseek: A new game for object detection and labeling. In
ECCV'12 Workshops 2012.
10
SLIDE 11
Motivation
?
11
SLIDE 12
Outline
- Motivation
- Object Segmentation
- Experiments
- Results
- Conclusions
- Next steps
12
SLIDE 13
Constrained parametric min-cuts for automatic object segmentation (CPMC)
- J. Carreira and C. Sminchisescu. Constrained parametric min-cuts for automatic object
- segmentation. In CVPR'10, 2010.
13
SLIDE 14
Constrained parametric min-cuts for automatic object segmentation
14
SLIDE 15
Motivation
CPMC
15
SLIDE 16
Outline
- Motivation
- Object Segmentation
- Experiments
- Results
- Conclusions
- Ongoing Work
16
SLIDE 17
Experiments
How many clicks do we need to achieve a certain quality in the segmentation? Test the algorithm for a large image dataset
17
SLIDE 18
Pascal VOC2010
1928 images divided in: Train (964) Validation (964)
18
SLIDE 19
Problem
Simulator
19
SLIDE 20
Simulator
- The simulator generates points using the ground truth of the image.
20
SLIDE 21
Simulator: Location of clicks
- S. Goferman, L. Zelnik-Manor, and
- A. Tal. Context-aware saliency
- detection. PAMI, 2012.
21
SLIDE 22
Simulator: Foreground/Background ratio
22
SLIDE 23
Outline
- Motivation
- Object Segmentation
- Experiments
- Results
- Conclusions
- Ongoing Work
23
SLIDE 24
Jaccard index
Measure of similarity between the segmentation result and the ground truth mask
24
SLIDE 25
Results
Using Pascal VOC2010 (Validation)
25
SLIDE 26
Results
Using Pascal VOC2010 (Validation)
26
SLIDE 27
Outline
- Motivation
- Object Segmentation
- Experiments
- Results
- Conclusions
- Ongoing Work
27
SLIDE 28
Conclusions
- Realistic simulator to process large amounts of data.
- Estimation of the expected AVERAGE Jaccard index by clicks.
- Inter-class variance of results.
28
SLIDE 29
Ongoing Work
29
SLIDE 30
Ongoing Work
- Image segmentation
- CPMC candidates
- Label propagation through
hierarchical partitions (eg. UCM, BPT…)
- Grabcut + Superpixels
30
SLIDE 31
Ongoing Work
- Data collection
- Awarded with $250 in CrowdMM Competition (ACM MM Barcelona 2013).
- Already more than 1500 games collected with 100 users
More on that in our poster!
31
SLIDE 32
Questions, suggestions…
Thank you for your attention
- Motivation
- Object Segmentation
- Experiments
- Results
- Conclusions
- Ongoing Work
32