Tentative schedule and topics

DAY 1: FEATURE LEARNING FOR FACE RECOGNITION & EVALUATION

FACE RECOGNITION (WEIHONG DENG)

·         Background, Pipeline and Historical Overview

·         Recent Progress and Trends

·         Basics and Importance of face alignment

FROM SUBSPACE to FEATURE LEARNING-1 (WEIHONG DENG)

·         Subspace learning: PCA versus LDA + hands-on

·         Basic techniques – revision of basic knowledge

·         Basic feature: LBP, HOG, and Gabor + hands-on

·         Discussion: Open problems in face recognition techniques

PERFORMANCE EVALUATION (NORMAN POH)

·         Evaluating biometric performance with hands-on

·         Experimental design of biometric systems

·         Confidence estimation + hands-on

INTRA-CLASS VARIATIONS OF BIOMETRIC DATA  (MOHAMMAD SABRI)

• Iranian Electronic National ID Card Program

 

DAY 2: METRIC LEARNING FOR FACE RECOGNITION

 FROM SUBSPACE to FEATURE LEARNING-2 (WEIHONG DENG)

·         Local feature learning: DFD, CBFD, PCANet

·         Hands-on exercise session 1

·         Deep feature learning: Deepface and DeepID

·         Relationship between Subspace and Feature learning

METRIC LEARNING (WEIHONG DENG)

·         Mahalanobis metric learning

·         Sparse reconstruction Metric Learning

·         Discriminative deep metric learning

·         Hands-on exercise session 2

·         Discussion: Open problems in learning techniques

 

 DAY 3: 3D FACE RECOGNITION

INTRODUCTION TO 3D FACE RECOGNITION (PAUL KOPPEN)

·         2D photos, stereo matching, 3D spatial coordinates

·         Photometric and geometric normalization

·         3D alignment and fitting – ICP, TPS, annotation, degrees of freedom

·         Hands-on exercise

·         3D registration, fixed topology

·         3D model construction – matrix representation, PCA

·         Hands-on exercise

ADVANCED TOPICS (PAUL KOPPEN)

·         3D Facial analysis – PCA, Non-negative Matrix Factorization

·         Automatic annotation – curvature, symmetry, geometric properties, viewpoint invariance

·         Open topics and challenges – discussion

 

DAY 4: AVANCED MACHINE LEARNING FOR FACE RECOGNITION AND FACE LIVENESS

ADVANCED MACHNE LEARNING TECHNIQUES (MANUEL GUNTHER)

·         Deep learning for face recognition

·         Parts-based face recognition

·         Hands-on (using Python or Matlab)

 

FACIAL ANTISPOOFING (NORMAN POH)

·         Facial anti spoofing techniques

·         Facial anti spoofing using Dynamic Mode Decomposition + hands-on