face detection and gender recognition ee368 project report michael bax chunlei liu ping li 28 may...
DESCRIPTION
RGB Colour-Space HistogramsTRANSCRIPT
![Page 1: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/1.jpg)
Face Detection Face Detection and Gender Recognitionand Gender Recognition
EE368 Project Report
Michael BaxChunlei Liu
Ping Li
28 May 2003
![Page 2: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/2.jpg)
Colour SpacesColour Spaces
![Page 3: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/3.jpg)
RGB Colour-Space HistogramsRGB Colour-Space Histograms
![Page 4: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/4.jpg)
HSV Colour-Space HistogramsHSV Colour-Space Histograms
![Page 5: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/5.jpg)
Empirical PDF ApproximationEmpirical PDF Approximation
![Page 6: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/6.jpg)
Pixel Classification Error (RGB)Pixel Classification Error (RGB)
![Page 7: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/7.jpg)
Pixel Classification Error (HSV)Pixel Classification Error (HSV)
![Page 8: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/8.jpg)
Input ImageInput Image
![Page 9: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/9.jpg)
Pixel Segmentation Pixel Segmentation Using the RGB Pixel PDFUsing the RGB Pixel PDF
![Page 10: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/10.jpg)
Non-Face Object RemovalNon-Face Object Removal
![Page 11: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/11.jpg)
Size-based Size-based Non-Face Object RemovalNon-Face Object Removal
![Page 12: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/12.jpg)
Location-based Location-based Non-Face Object RemovalNon-Face Object Removal
![Page 13: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/13.jpg)
Object Size Threshold CorrectionObject Size Threshold Correction
![Page 14: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/14.jpg)
PCA-basedPCA-basedNon-Face Object RemovalNon-Face Object Removal
![Page 15: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/15.jpg)
Connected Component Connected Component AnalysisAnalysis
Low pass filtering, hole filling and background rejection
Identification of connected faces based on statistical analysis
Iterative separation of connected regions
Preprocessing
Connected faces identification
Face separation
![Page 16: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/16.jpg)
Connected ComponentsConnected Components
![Page 17: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/17.jpg)
Component SeparationComponent Separation
![Page 18: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/18.jpg)
Separated ComponentsSeparated Components
![Page 19: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/19.jpg)
Component IdentificationComponent Identification
Template matching and peak thresholding to remove remaining non-face objects
Removal of repeated faces segments using a distance constraint
![Page 20: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/20.jpg)
Face Position RefinementFace Position Refinement
The face centre is located at the bridge of the nose
The centroid of the segmented face is somewhat inaccurate in finding face centres
Multi-scale, high threshold template matching finds centres more accurately
Use centroid for remaining faces
![Page 21: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/21.jpg)
Image Pyramid-based Image Pyramid-based Template MatchingTemplate Matching
Training face preprocessing– Training faces were rotation compensated,
registered, and resampled in greyscale– Resampled faces were averaged and masked
Greyscale input image pyramid composition– 20% scale increments
Normalized cross-correlation with nose bridge-centred average face template
![Page 22: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/22.jpg)
Finding Faces Finding Faces with Template Matchingwith Template Matching
High threshold for accurate centre location
Moderate threshold for robust backup face location – if morphological
subsystem gives unexpected results
![Page 23: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/23.jpg)
Gender DetectionGender Detection
Mean intensity Template matching
using average of each female face
Biased towards missing female faces to avoid false-positive penalty (9:1)
![Page 24: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/24.jpg)
Face Detection ResultsFace Detection Results
![Page 25: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/25.jpg)
Image Hits Repeated False Hits Distance Time (s) Bonus
1 21 0 0 11.1 91 2
2 24 0 0 15.6 90 2
3 25 0 0 10.5 97 0
4 24 0 0 11.8 97 1
5 24 0 0 10.7 103 0
6 24 0 0 9.6 94 0
7 22 0 0 11.2 88 1
Average 23.4 0 0 11.5 94 0.86
Results StatisticsResults Statistics
![Page 26: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/26.jpg)
![Page 27: Face Detection and Gender Recognition EE368 Project Report Michael Bax Chunlei Liu Ping Li 28 May 2003](https://reader036.vdocuments.site/reader036/viewer/2022062413/5a4d1b237f8b9ab0599960ff/html5/thumbnails/27.jpg)
17
16
15
14
13
12
11
10
2 (1)1 (22)9
1 (2)8 (14)8
4 (0)7 (16)7
4 (0)5 (17)6
4 (0)9 (13)5
4 (0)4 (18)4
4 (0)4 (18)3
4 (0)2 (20)2
2 (1)3 (19)1
Gender RecognitionFace Detection
17
16
15
14
13
12
11
10
2 (1)1 (22)9
1 (2)8 (14)8
4 (0)7 (16)7
4 (0)5 (17)6
4 (0)9 (13)5
4 (0)4 (18)4
4 (0)4 (18)3
4 (0)2 (20)2
2 (1)3 (19)1
Gender RecognitionFace Detection