Skin
detection by dual maximization of detectors agreement for video monitoring
[Description][Related publication][Dataset][Sample results][Software][Additional references]
This page presents
sample results of our approach for skin detection which is able to adapt its
parameters to image data and select the optimum color space channels for
maximizing performance. Results on human activity recognition datasets are
compared against state-of-the-art approaches.
Evaluation data with ground truth data and a MATLAB
implementation are also provided.
Skin detection by dual maximization of detectors
agreement for video monitoring [preprint pdf]
J. SanMiguel and S. Suja
Pattern Recognition Letters, 34(16):2102–2109, Dec. 2013
Contact Information
Juan C. SanMiguel - Show email
The evaluation set is a
composition of several datasets. The following describes the details and
provides the links for downloading:
|
Set |
|
SSG |
||||
|
Train |
#images |
60 |
30 |
25 |
25 |
25 |
|
|
#total pixels |
4608000 |
12441600 |
4320000 |
8640000 |
2534400 |
|
|
#skin pixels |
23376 |
272219 |
61818 |
50826 |
73264 |
|
|
|
|||||
|
|
|
|
|
|
|
|
|
|
|
|||||
|
|
|
|
|
|
|
|
|
Test |
#images |
25 |
25 |
25 |
25 |
25 |
|
|
#total pixels |
2304000 |
10368000 |
4320000 |
8640000 |
2534400 |
|
|
#skin pixels |
10816 |
191595 |
85036 |
40983 |
60507 |
|
|
|
|||||
|
|
|
|
|
|
|
|
|
|
|
|||||
|
|
|
|
||||
Some sample frames are
depicted in the following figure:
|
|
Select the directories that contain image samples from
the links below to look at the sample experimental results of our proposed
approach.
|
Proposed approach [Summary of all results] ·
EDds: [Slideshow] [PNG
files] [Summary] ·
LIRIS: [Slideshow] [PNG files] [Summary] ·
SSG: [Slideshow] [PNG
files] [Summary] ·
UT: [Slideshow] [PNG files] [Summary] ·
AMI: [Slideshow] [PNG files] [Summary] Comparative results [summary] ·
EDds: [Slideshow] [PNG
files] ·
LIRIS: [Slideshow] [PNG files] |
|
The
software listed in this section can be freely used for research purposes (also available in GitHub):
- SDDMA_adaptive_skin_detector
v1.0 (download
link) (light dataset – 7.8MB)
- SDDMA_adaptive_skin_detector_full
v1.0 (download
link) (full dataset – 152MB)
References
[T_HS2001] Wang, Y., Yuan, B., 2001. A novel approach
for human face detection from color images under complex background. Pattern
Recognition 34 (10), 1983-1992.
[T_CbCr2001] Wang, Y., Yuan, B., 2001. A novel
approach for human face detection from color images under complex background.
Pattern Recognition 34 (10), 1983-1992.
[BAY2002]
Jones, M., Rehg, J., 2002. Statistical color
models with application to skin detection. Int. Journal of Computer Vision 46
(1), 81-96.
[ASD2006]
Dadgostar, F., Sarrafzadeh, A., 2006. An adaptive
real-time skin detector based on hue thresholding: A comparison on two motion
tracking methods. Pattern Recognition Letters 27 (12), 1342-1352.
[MMI2007] Conaire, C., O'Connor, N., Smeaton, A.,
2007. Detector adaptation by maximising agreement between independent data
sources. In: IEEE Int. Conf. on Computer Vision and Pattern Recognition (CVPR).
pp. 1-6.
[RF2010] Khan, R., Hanbury, A., Stoettinger, J., 2010.
Skin detection: A random forest approach. In: IEEE Int. Conf. on Image
Processing (ICIP). pp. 4613-4616.