AI helps you reading Science

AI generates interpretation videos

AI extracts and analyses the key points of the paper to generate videos automatically


pub
Go Generating

AI Traceability

AI parses the academic lineage of this thesis


Master Reading Tree
Generate MRT

AI Insight

AI extracts a summary of this paper


Weibo:
AdaBoost classifier is used with Haar and Local Binary Pattern features whereas Support Vector Machine classifier is used with Histogram of Oriented Gradients features for face detection evaluation

Image-based Face Detection and Recognition: "State of the Art"

CoRR, (2013)

Cited by: 92|Views6
EI
Full Text
Bibtex
Weibo

Abstract

Face recognition from image or video is a popular topic in biometrics research. Many public places usually have surveillance cameras for video capture and these cameras have their significant value for security purpose. It is widely acknowledged that the face recognition have played an important role in surveillance system as it doesn't...More

Code:

Data:

0
Introduction
  • Over the last few decade lots of work is been done in face detection and recognition [14] as it’s a best way for person identification [16] because it doesn’t require human cooperation [15] so that it became a hot topic in biometrics.
  • In current paper the authors developed a system for the said method’s evaluation as a first milestone for video based face detection and recognition for surveillance.
Highlights
  • Over the last few decade lots of work is been done in face detection and recognition [14] as it’s a best way for person identification [16] because it doesn’t require human cooperation [15] so that it became a hot topic in biometrics
  • The following paper discuss about face detection methods in section II, section III discuss about face recognition methods based on the results of section II; result summery has been provided in the form of tables
  • AdaBoost [6] classifier is used with Haar [7] and Local Binary Pattern (LBP) [8] features whereas Support Vector Machine (SVM) [12] classifier is used with Histogram of Oriented Gradients (HOG) [13] features for face detection evaluation
  • In current work we developed the system to evaluate the face detection and recognition methods which are
Results
  • The following paper discuss about face detection methods in section II, section III discuss about face recognition methods based on the results of section II; result summery has been provided in the form of tables.
  • AdaBoost [6] classifier is used with Haar [7] and Local Binary Pattern (LBP) [8] features whereas Support Vector Machine (SVM) [12] classifier is used with Histogram of Oriented Gradients (HOG) [13] features for face detection evaluation.
  • To reduce pose variation and illumination in extracted faces two extra actions performed in pre-processing stage to improve recognition results: 1) Eyes detection is been used to remove head turn, tilt, slant and position of face, demonstrated in figure 4; 2) Histogram equalization is been performed.
  • In dataset [1], face collection with plain green background; no head scale and light variation but having minor changes in head turn, tilt, slant, position of face and considerable change in expressions.
  • In dataset [2], face collection with red curtain background, variation is caused by shadows as subject moves forward, having minor changes in head turn, tilt and slant; large head scale variation; some expression variation, translation in position of face and image lighting variation as subject moves forward, significant lighting changes occur on faces moment due to the artificial lighting arrangement.
  • In dataset [3], face collection with complex background; large head scale variation; minor variations in head turn, tilt, slant and expression; some translation in face position and significant light variation because of object moment in artificial light.
  • In dataset [4], face collection with plain background; small head scale variation; considerable variation in head turn, tilt, slant and major variation in expression; minor translation in face position and light variation.
  • In dataset [5], face collection with constant background having minor head scale variation and light variation; huge variation in turn, tilt, slant, expression and face position.
Conclusion
  • In current work the authors developed the system to evaluate the face detection and recognition methods which are
  • In current system Haar-like [7] features reported relatively well but it has much false detection than LBP [8] which could be consider being a future work in surveillance to reduce false detection in Haar-like [7] features and for the recognition part gabor [11] is reported well as it’s qualities overcomes datasets complexity.
Summary
  • Over the last few decade lots of work is been done in face detection and recognition [14] as it’s a best way for person identification [16] because it doesn’t require human cooperation [15] so that it became a hot topic in biometrics.
  • In current paper the authors developed a system for the said method’s evaluation as a first milestone for video based face detection and recognition for surveillance.
  • The following paper discuss about face detection methods in section II, section III discuss about face recognition methods based on the results of section II; result summery has been provided in the form of tables.
  • AdaBoost [6] classifier is used with Haar [7] and Local Binary Pattern (LBP) [8] features whereas Support Vector Machine (SVM) [12] classifier is used with Histogram of Oriented Gradients (HOG) [13] features for face detection evaluation.
  • To reduce pose variation and illumination in extracted faces two extra actions performed in pre-processing stage to improve recognition results: 1) Eyes detection is been used to remove head turn, tilt, slant and position of face, demonstrated in figure 4; 2) Histogram equalization is been performed.
  • In dataset [1], face collection with plain green background; no head scale and light variation but having minor changes in head turn, tilt, slant, position of face and considerable change in expressions.
  • In dataset [2], face collection with red curtain background, variation is caused by shadows as subject moves forward, having minor changes in head turn, tilt and slant; large head scale variation; some expression variation, translation in position of face and image lighting variation as subject moves forward, significant lighting changes occur on faces moment due to the artificial lighting arrangement.
  • In dataset [3], face collection with complex background; large head scale variation; minor variations in head turn, tilt, slant and expression; some translation in face position and significant light variation because of object moment in artificial light.
  • In dataset [4], face collection with plain background; small head scale variation; considerable variation in head turn, tilt, slant and major variation in expression; minor translation in face position and light variation.
  • In dataset [5], face collection with constant background having minor head scale variation and light variation; huge variation in turn, tilt, slant, expression and face position.
  • In current work the authors developed the system to evaluate the face detection and recognition methods which are
  • In current system Haar-like [7] features reported relatively well but it has much false detection than LBP [8] which could be consider being a future work in surveillance to reduce false detection in Haar-like [7] features and for the recognition part gabor [11] is reported well as it’s qualities overcomes datasets complexity.
Tables
  • Table1: Face detection results summery Detection
  • Table2: Face recognition results summery
  • Table3: Face database summery
Download tables as Excel
Study subjects and analysis
datasets: 5
Considering these devastating capacities and its great success in face recognition Gabor [11] features are insensitive to transformations as. Five datasets been used for above experiments. In dataset [1], face collection with plain green background; no head scale and light variation but having minor changes in head turn, tilt, slant, position of face and considerable change in expressions

datasets: 5
Image/Individual ~20 20 ~20 20 26 considered to be a bench mark. Some methods performed consistently over different datasets whereas other methods behave very randomly however based on average experimental results performance is evaluated, five datasets been used for this purpose. Face detection and recognition method’s result summery is provided in table 1 and table 2 respectively whereas datasets summery is provided in table 3

Reference
  • Face Recognition Data, University of Essex, UK, Face 94, http://cswww.essex.ac.uk/mv/all faces/faces94.html.
    Findings
  • Face Recognition Data, University of Essex, UK, Face 95, http://cswww.essex.ac.uk/mv/all faces/faces95.html.
    Findings
  • Face Recognition Data, University of Essex, UK, Face 96, http://cswww.essex.ac.uk/mv/all faces/faces96.html.
    Findings
  • Face Recognition Data, University of Essex, UK, Grimace, http://cswww.essex.ac.uk/mv/all faces/grimace.html.
    Findings
  • Psychological Image Collection at Stirling (PICS), Pain Expressions, http://pics.psych.stir.ac.uk/2D_face_sets.htm.
    Findings
  • K. T. Talele, S. Kadam, A. Tikare, Efficient Face Detection using Adaboost, “IJCA Proc on International Conference in Computational Intelligence”, 2012.
    Google ScholarLocate open access versionFindings
  • [8] T. Ahonen, A. Hadid, M. Peitikainen, Face recognition with local binary patterns. “In Proc. of European Conference of Computer Vision”, 2004.
    Google ScholarLocate open access versionFindings
  • [9] M. A. Turk and A.P. Pentland, Face recognition using eigenfaces, “Proceedings of the IEEE”, 586-591, 1991.
    Google ScholarLocate open access versionFindings
  • [10] J Lu, K. N. Plataniotis, A. N. Venetsanopoulos, Face recognition using LDA-based algorithms, “IEEE Neural Networks Transaction”, 2003.
    Google ScholarFindings
  • [11] L. Wiskott, M. Fellous, N. Krger, and C. Malsburg, Face recognition by elastic bunch graph matching, “IEEE Trans”, on PAMI, 19:775–779, 1997.
    Google ScholarLocate open access versionFindings
  • [12] I. Kukenys, B. McCane, Support Vector Machines for Human Face Detection, “Proceedings of the New Zealand Computer Science Research Student Conference”, 2008.
    Google ScholarLocate open access versionFindings
  • [13] M. M. Abdelwahab, S. A. Aly, I. Yousry, Efficient WebBased Facial Recognition System Employing 2DHOG, arXiv:1202.2449v1 [cs.CV].
    Findings
  • [14] W. Zhao, R. chellappa, P. J. Phillips, Face recognition: A literature survey, “ACM Computing Surveys (CSUR)”, December 2003.
    Google ScholarFindings
  • [15] G. L. Marcialis, F. Roli, Chapter: Fusion of Face Recognition Algorithms for Video-Based Surveillance Systems, Department of Electrical and Electronic Engineering- Univ- ersity of Cagliari- Italy.
    Google ScholarFindings
  • [16] A. Suman, Automated face recognition: Applications within law enforcement. Market and technology review, “NPIA”, 2006.
    Google ScholarFindings
Your rating :
0

 

Tags
Comments
小科