Advertisement
Advertisement

More Related Content

Slideshows for you(17)

Advertisement

Similar to Sparse Support Faces - Battista Biggio - Int'l Conf. Biometrics, ICB 2015, Phuket, Thailand, May 19-22, 2015(20)

Advertisement

Sparse Support Faces - Battista Biggio - Int'l Conf. Biometrics, ICB 2015, Phuket, Thailand, May 19-22, 2015

  1. Pa#ern  Recogni-on     and  Applica-ons  Lab                                     University   of  Cagliari,  Italy     Department  of   Electrical  and  Electronic   Engineering   Sparse Support Faces Ba#sta  Biggio,  Marco  Melis,  Giorgio  Fumera,  Fabio  Roli         Dept.  Of  Electrical  and  Electronic  Engineering   University  of  Cagliari,  Italy   Phuket,  Thailand,  May  19-­‐22,  2015  ICB  2015  
  2.   http://pralab.diee.unica.it Template-based Face Verification 2   gc ≥ϑc genuine   impostor   true   false   s(x,tc i ){ }i=1 p Matcher    s(⋅,⋅) Fusion   rule   gc (x)xFeature   extrac-on   Verifica-on  is  based  on  how  similar  the  submi#ed  image  is  to  the  client’s  templates   Client-­‐specific  one-­‐class  classifica:on   mean gc (x) = 1 p s(x,tc i ) i=1 p ∑ gc (x) = max i=1,…,p s(x,tc i )max Claimed   Iden-ty   tc 1 , …, tc p { } Claimed  iden-ty’s  templates  
  3.   http://pralab.diee.unica.it Cohort-based Face Verification 3   Verifica-on  is  based  on  how  similar  the  submi#ed  image  is  to  the  client’s  templates   and  on  how  different  it  is  from  the  cohorts’  templates   Client-­‐specific  two-­‐class  classifica:on  (one-­‐vs-­‐all)   gc ≥ϑc genuine   impostor   true   false   s(x,tc i ){ }i=1 n Matcher    s(⋅,⋅) Fusion   rule   gc (x)xFeature   extrac-on   tc 1 , …, tc p { } Claimed  iden-ty’s  templates   Cohorts   tc p+1 , …, tc n { } Claimed   Iden-ty  
  4.   http://pralab.diee.unica.it Cohort-based Fusion Rules •  Cohort selection is heuristically driven –  e.g., selection of the closest cohorts to the client’s templates •  Cohort-based fusion rules are also based on heuristics –  Test-normalization [Auckenthaler et al., DSP 2000] –  Aggarwal’s max rule [Aggarwal et al., CVPR-W 2006] 4   gc (x) = 1 σc (x) 1 p s(x,tc i ) i=1 p ∑ −µc (x) # $ % & ' ( gc (x) = max i=1,…,p s(x,tc i ) max j=p+1,…,n s(x,tc j )
  5.   http://pralab.diee.unica.it Open Issues •  Fusion rules and cohort selection are based on heuristics –  No guarantees of optimality in terms of verification error •  Our goal: to design a procedure to optimally select the reference templates and the fusion rule –  Optimal in the sense that it minimizes verification error (FRR and FAR) •  Underlying idea: to consider face verification as a two-class classification problem in similarity space 5  
  6.   http://pralab.diee.unica.it s(x, ) s(x, ) Face Verification in Similarity Space •  The matching function maps faces onto a similarity space –  How to design an optimal decision function in this space? 6   ?  
  7.   http://pralab.diee.unica.it Support Face Machines (SFMs) •  We learn a two-class SVM for each client –  using the matching score as the kernel function –  genuine client y=+1, impostors y=-1 •  SVM minimizes the classification error (optimal in that sense) –  FRR and FAR in our case •  The fusion rule is a linear combination of matching scores •  The templates are automatically selected for each client –  support vectors à support faces 7   gc (x) = αis(x,tc i ) i ∑ − αjs(x,tc j ) j ∑ + b
  8.   http://pralab.diee.unica.it Support Face Machines (SFMs) 8   s(x, ) s(x, ) •  Maximum-margin classifiers gc (x) = αis(x,tc i ) i ∑ − αjs(x,tc j ) j ∑ + b
  9.   http://pralab.diee.unica.it Sparse Support Faces •  Open issue: SFMs require too many support faces –  Number of support faces scales linearly with training set size •  Our goal: to learn a much sparser combination of match scores •  by jointly optimizing the weighting coefficients and support faces: 9   hc (x) = βis(x, zc k )+ b k=1 m ∑ , m << n min β,z Ω β, z( )= 1 n uk gc (xk )− hc (xk )( ) 2 + λβT β i=1 n ∑
  10.   http://pralab.diee.unica.it z-­‐step Sparse Support Faces 10   SFM with 12 support faces −5 0 5 −5 0 5 −5 0 5 SSFM with 4 virtual faces −5 0 5 −5 0 5 −5 0 5 β-­‐step   Solu:on  algorithm  is  an  itera-ve  two-­‐step  procedure:   If s(x,z) is not differentiable or analytically given, gradient can be approximated    
  11.   http://pralab.diee.unica.it 0.5 1 2 5 10 0 5 10 15 20 AT&T − RBF Kernel FAR (%) FRR(%) mean (5) max (5) t−norm (10) aggarwal−max (10) SFM (37.5 ± 3.8) SFM−sel (10) SFM−red (2) SSFM (2) Experiments 11   Datasets: AT&T (40 clients, 10 images each) BioID (23 clients, 1,521 images) Matcher: PCA+RBF kernel (exact gradient) 5 repetitions, different clients in TR/TS splits TR: 5 images/client 0.5 1 2 5 10 0 10 20 30 40 BioID − RBF Kernel FAR (%) FRR(%) mean (5) max (5) t−norm (10) aggarwal−max (10) SFM (23.9 ± 2.7) SFM−sel (10) SFM−red (2) SSFM (2)
  12.   http://pralab.diee.unica.it Experiments 12  0.5 1 2 5 10 0 10 20 30 40 BioID − EBGM FAR (%) FRR(%) mean (5) max (5) t−norm (10) aggarwal−max (10) SFM (15.0 ± 2.6) SFM−sel (5) SFM−red (5) SSFM (5) 0.5 1 2 5 10 0 5 10 15 20 AT&T − EBGM FAR (%) FRR(%) mean (5) max (5) t−norm (10) aggarwal−max (10) SFM (19.5 ± 3.0) SFM−sel (5) SFM−red (5) SSFM (5) Datasets: AT&T (40 clients, 10 images each) BioID (23 clients, 1,521 images) Matcher: EBGM (approx. gradient) 5 repetitions, different clients in TR/TS splits TR: 5 images/client
  13.   http://pralab.diee.unica.it From Support Faces to Sparse Support Faces •  A client’s gallery of 17 support faces (and weights) reduced to 5 virtual templates by our sparse support face machine –  Dataset: BioID –  Matching algorithm: EBGM 13   4.040 2.854 −0.997 −3.525 −2.208
  14.   http://pralab.diee.unica.it Conclusions and Future Research Directions •  Sparse support face machines: –  reduce computational time and storing requirements during verification without affecting verification accuracy –  by jointly learning an optimal combination of matching scores, and a corresponding sparse set of virtual support faces •  No explicit feature representation is required –  Matching algorithm exploited as kernel function –  Virtual templates created exploiting approximations of its gradient •  Future work –  Fingerprint verification –  Identification setting •  Joint reduction of virtual templates for each client-specific classifier 14  
  15.   http://pralab.diee.unica.it ?  Any questions Thanks  for  your  a#en-on!   15   Code available at: http://pralab.diee.unica.it/en/SSFCodeProject
Advertisement