Robust Face Recognition under Varying Illumination and Occlusion - - PowerPoint PPT Presentation

▶
robust face recognition under varying illumination and
SMART_READER_LITE
LIVE PREVIEW

Robust Face Recognition under Varying Illumination and Occlusion - - PowerPoint PPT Presentation

Jul 09, 2023 830 likes •1.08k views

Robust Face Recognition under Varying Illumination and Occlusion Considering Structured Sparsity Xingjie Wei, Chang-Tsun Li and Yongjian Hu Department of Computer Science, University of Warwick x.wei@warwick.ac.uk http://warwick.ac.uk/xwei

slide-1
SLIDE 1

Xingjie Wei, Chang-Tsun Li and Yongjian Hu

Department of Computer Science, University of Warwick

x.wei@warwick.ac.uk http://warwick.ac.uk/xwei

Robust Face Recognition under Varying Illumination and Occlusion Considering Structured Sparsity

slide-2
SLIDE 2

Face

  • People love faces !

– Biological nature – Sensitive to the face pattern

2

A house with a Hitler face

slide-3
SLIDE 3

Face Recognition

  • Uncontrolled conditions: large changes in

pose, illumination, expression and occlusion, aging… Still challenging

3

slide-4
SLIDE 4

4

Motivation

  • Face recognition in real-world environments
  • ften has to confront with uncontrolled and

uncooperative conditions

– illumination changes, occlusion

  • Uncontrolled variations are usually coupled
  • Less work focuses on simultaneously

handling them

slide-5
SLIDE 5

5

Our Method

  • Our work deals with the illumination changes

and occlusion simultaneously considering structured sparsity

Sparse Representation flat sparsity

represents a test image using minimal number of training images from all classes represents a test image using the minimal number of clusters

slide-6
SLIDE 6

6

Our Method

  • Our work deals with the illumination changes

and occlusion simultaneously considering structured sparsity aided with:

– Structural occlusion dictionary: better modelling contiguous occlusion

contiguous occlusion also forms a cluster structure

slide-7
SLIDE 7

7

Our Method

  • Our work deals with the illumination changes

and occlusion simultaneously considering structured sparsity aided with:

– Structural occlusion dictionary: better modelling contiguous occlusion – WLD feature: robust to illumination changes, remove shadows Inspired by the psychophysical Weber’s Law

slide-8
SLIDE 8

Sparse Representation

8

  • Models a test image as a linear

combination of training images

– Using minimal number of training images

y X α = ×

1 1 . . .

… sparse

slide-9
SLIDE 9

Sparse Representation

  • Involves training images from all classes

– Optimal for reconstruction but not necessary for classification

9

Using the same number of base vectors

slide-10
SLIDE 10

10

Our Method

  • Structured Sparsity

– Each class form a cluster

cluster structure

slide-11
SLIDE 11

11

Our Method

  • Structured Sparsity

– Represents a test image using the minimum number of clusters

y X α = ×

1 1 . .

… …

cluster1 cluster2

…

slide-12
SLIDE 12

Sparse Representation

  • Occlusion modelling: identity matrix

– limitation: is able to represent any image of size m

12

× =

1 1 . . 1

… … I sparse size: m X

m base vectors

slide-13
SLIDE 13

Our method

  • Contiguous occlusion: the nonzeros entries

are likely to be spatially continuous, are aligned to clusters

13

index of occlusion base vectors size: 83*60=4980

slide-14
SLIDE 14

Our method

  • Structural occlusion dictionary

– uses the cluster occlusion dictionary to replace the identity matrix I

14

× =

1 1 … 1 1

… … D

cluster2 cluster (s+1)

…

cluster1

…

cluster s

X cluster structure

slide-15
SLIDE 15

15

Our Method

  • Extreme illumination + occlusion:

– coupled occlusion takes up a large ratio of the image – not “sparse” error

slide-16
SLIDE 16

16

Our Method

  • A different view: extract relevant features

that reduce the difference

  • Using WLD feature

+1 +1 +1 +1

  • 8

+1 +1 +1 +1 +1

Filtering Original image WLD feature

 Maintain most salient facial features  Insensitive to illumination changes  Can correct shadow effects

Chen et al, Wld: A robust local image descriptor, PAMI, 2010

slide-17
SLIDE 17

Illustrative Example

17 Test image Estimated

  • cclusion

Reconstruction Sparse coefficients Residuals Reference image Sparse coefficients Residuals Test image Estimated

  • cclusion

Reconstruction Reference image

belongs to class 1 class 1

slide-18
SLIDE 18

Experiments

  • Synthetic Occlusion with Extreme

Illumination

– Extended Yale B database – Occlusion levels: 0% ~ 50% of the image

18

Subset 3 Subset 4 Subset 5 Training set Testing set

slide-19
SLIDE 19

Experiments

  • Synthetic Occlusion with Extreme

Illumination

– using only the raw pixel intensity as feature

19

[15] Wright et al, TPAMI, 2009. [17] Zhang et al, ICCV, 2011

slide-20
SLIDE 20

Experiments

  • Synthetic Occlusion with Extreme

Illumination

– using WLD feature

20

[15] Wright et al, TPAMI, 2009. [16] Yang et al, ECCV, 2010

slide-21
SLIDE 21

Experiments

  • Synthetic Occlusion with Extreme

Illumination

– using WLD feature

21

[15] Wright et al, TPAMI, 2009. [16] Yang et al, ECCV, 2010

slide-22
SLIDE 22

Experiments

  • Disguise with Non-uniform Illumination

– The AR Database – Real occlusion, 2 sessions

22

Training set Testing set

slide-23
SLIDE 23

Experiments

  • Disguise with Non-uniform Illumination

23

slide-24
SLIDE 24

24

Thank you

  • Questions ?
  • Xingjie Wei
  • x.wei@warwick.ac.uk
  • http://warwick.ac.uk/xwei
  • Department of Computer Science, University
  • f Warwick