Challenges in Learning the Appearance of Faces for Automated Image Analysis: part I alessandro verri DISI – università di genova

Slides:



Advertisements
Similar presentations
Real-Time Detection, Alignment and Recognition of Human Faces
Advertisements

Active Appearance Models
Object Detection Using Semi- Naïve Bayes to Model Sparse Structure Henry Schneiderman Robotics Institute Carnegie Mellon University.
Detecting Faces in Images: A Survey
Rapid Object Detection using a Boosted Cascade of Simple Features Paul Viola, Michael Jones Conference on Computer Vision and Pattern Recognition 2001.
Rapid Object Detection using a Boosted Cascade of Simple Features Paul Viola, Michael Jones Conference on Computer Vision and Pattern Recognition 2001.
Face detection Behold a state-of-the-art face detector! (Courtesy Boris Babenko)Boris Babenko.
Face Detection & Synthesis using 3D Models & OpenCV Learning Bit by Bit Don Miller ITP, Spring 2010.
An Introduction to Face Detection and Recognition
Performance Evaluation Measures for Face Detection Algorithms Prag Sharma, Richard B. Reilly DSP Research Group, Department of Electronic and Electrical.
AdaBoost & Its Applications
Face detection Many slides adapted from P. Viola.
Detecting Pedestrians by Learning Shapelet Features
The Viola/Jones Face Detector Prepared with figures taken from “Robust real-time object detection” CRL 2001/01, February 2001.
The Viola/Jones Face Detector (2001)
Robust Moving Object Detection & Categorization using self- improving classifiers Omar Javed, Saad Ali & Mubarak Shah.
Rapid Object Detection using a Boosted Cascade of Simple Features
Object Recognition with Informative Features and Linear Classification Authors: Vidal-Naquet & Ullman Presenter: David Bradley.
Robust Real-time Object Detection by Paul Viola and Michael Jones ICCV 2001 Workshop on Statistical and Computation Theories of Vision Presentation by.
Face detection and recognition Many slides adapted from K. Grauman and D. Lowe.
A Robust Real Time Face Detection. Outline  AdaBoost – Learning Algorithm  Face Detection in real life  Using AdaBoost for Face Detection  Improvements.
Face Recognition with Harr Transforms and SVMs EE645 Final Project May 11, 2005 J Stautzenberger.
4EyesFace-Realtime face detection, tracking, alignment and recognition Changbo Hu, Rogerio Feris and Matthew Turk.
CS 223B Assignment 1 Help Session Dan Maynes-Aminzade.
A Robust Real Time Face Detection. Outline  AdaBoost – Learning Algorithm  Face Detection in real life  Using AdaBoost for Face Detection  Improvements.
Robust Real-Time Object Detection Paul Viola & Michael Jones.
Face Processing System Presented by: Harvest Jang Group meeting Fall 2002.
Dynamic Cascades for Face Detection 第三組 馮堃齊、莊以暘. 2009/01/072 Outline Introduction Dynamic Cascade Boosting with a Bayesian Stump Experiments Conclusion.
Object Detection Using the Statistics of Parts Henry Schneiderman Takeo Kanade Presented by : Sameer Shirdhonkar December 11, 2003.
A Brief Survey on Face Recognition Systems Amir Omidvarnia March 2007.
Face Detection and Recognition
Foundations of Computer Vision Rapid object / face detection using a Boosted Cascade of Simple features Presented by Christos Stoilas Rapid object / face.
Face Detection CSE 576. Face detection State-of-the-art face detection demo (Courtesy Boris Babenko)Boris Babenko.
FACE DETECTION AND RECOGNITION By: Paranjith Singh Lohiya Ravi Babu Lavu.
Facial Feature Detection
Face Detection using the Viola-Jones Method
EADS DS / SDC LTIS Page 1 7 th CNES/DLR Workshop on Information Extraction and Scene Understanding for Meter Resolution Image – 29/03/07 - Oberpfaffenhofen.
Introduction to Computer Vision Olac Fuentes Computer Science Department University of Texas at El Paso El Paso, TX, U.S.A.
“Secret” of Object Detection Zheng Wu (Summer intern in MSRNE) Sep. 3, 2010 Joint work with Ce Liu (MSRNE) William T. Freeman (MIT) Adam Kalai (MSRNE)
Object Detection Using the Statistics of Parts Presented by Nicholas Chan – Advanced Perception Robust Real-time Object Detection Henry Schneiderman.
Window-based models for generic object detection Mei-Chen Yeh 04/24/2012.
Face detection Slides adapted Grauman & Liebe’s tutorial
Object Recognition in Images Slides originally created by Bernd Heisele.
Robust Real-time Face Detection by Paul Viola and Michael Jones, 2002 Presentation by Kostantina Palla & Alfredo Kalaitzis School of Informatics University.
ECE738 Advanced Image Processing Face Detection IEEE Trans. PAMI, July 1997.
Face Recognition: An Introduction
Face Detection Ying Wu Electrical and Computer Engineering Northwestern University, Evanston, IL
Face Detection Using Large Margin Classifiers Ming-Hsuan Yang Dan Roth Narendra Ahuja Presented by Kiang “Sean” Zhou Beckman Institute University of Illinois.
Real-Time Detection, Alignment and Recognition of Human Faces Rogerio Schmidt Feris Changbo Hu Matthew Turk Pattern Recognition Project June 12, 2003.
Robust Real Time Face Detection
Adaboost and Object Detection Xu and Arun. Principle of Adaboost Three cobblers with their wits combined equal Zhuge Liang the master mind. Failure is.
HCI/ComS 575X: Computational Perception Instructor: Alexander Stoytchev
Categorization by Learning and Combing Object Parts B. Heisele, T. Serre, M. Pontil, T. Vetter, T. Poggio. Presented by Manish Jethwa.
Lecture 09 03/01/2012 Shai Avidan הבהרה: החומר המחייב הוא החומר הנלמד בכיתה ולא זה המופיע / לא מופיע במצגת.
The Viola/Jones Face Detector A “paradigmatic” method for real-time object detection Training is slow, but detection is very fast Key ideas Integral images.
FACE DETECTION : AMIT BHAMARE. WHAT IS FACE DETECTION ? Face detection is computer based technology which detect the face in digital image. Trivial task.
Notes on HW 1 grading I gave full credit as long as you gave a description, confusion matrix, and working code Many people’s descriptions were quite short.
Project Overview CSE 6367 – Computer Vision Vassilis Athitsos University of Texas at Arlington.
Face Detection and Head Tracking Ying Wu Electrical Engineering & Computer Science Northwestern University, Evanston, IL
3D Face Recognition Using Range Images Literature Survey Joonsoo Lee 3/10/05.
Face detection Many slides adapted from P. Viola.
Face Detection 蔡宇軒.
CONTENTS:  Introduction.  Face recognition task.  Image preprocessing.  Template Extraction and Normalization.  Template Correlation with image database.
Reading: R. Schapire, A brief introduction to boosting
Guillaume-Alexandre Bilodeau
Presented by Minh Hoai Nguyen Date: 28 March 2007
High-Level Vision Face Detection.
Learning to Detect Faces Rapidly and Robustly
Cos 429: Face Detection (Part 2) Viola-Jones and AdaBoost Guest Instructor: Andras Ferencz (Your Regular Instructor: Fei-Fei Li) Thanks to Fei-Fei.
Categorization by Learning and Combing Object Parts
Presentation transcript:

Challenges in Learning the Appearance of Faces for Automated Image Analysis: part I alessandro verri DISI – università di genova

actually, i’m gonna talk about: ► brief introduction (the whole thing) ► what some people do for detecting faces ► what we are doing

the problem(s) ► geometry (position, rotation, pose, scale,…) ► facial features (beards, glasses,…) ► facial expressions ► occlusions ► imaging conditions (illumination, resolution, color contrast, camera quality,…)

where we are: face detection we address face detection as a brute force classification problem (sometimes sophisticated, but still brute force) the model is encoded in the training samples but not explicitly defined

face recognition and the like explicit image models are derived from examples separating identity and imaging parameters

motivation we want to explore who should learn from whom… we come back to this at the end!

some approaches ► knowledge-based (Yang & Huang 94) ► feature invariant (Leung et al, 95; Yow & Cipolla, 97) ► template matching (Lanitis et al, 95) ► appearance based eigenfaces (Turk & Pentland, 91)eigenfaces (Turk & Pentland, 91) SVM (Osuna et al, 97)SVM (Osuna et al, 97) naive bayes (Schneiderman & Kanade, 98)naive bayes (Schneiderman & Kanade, 98) AdaBoost (Viola and Jones, 01)AdaBoost (Viola and Jones, 01)

SVM: global detector (Poggio’s group) ► some preprocessing essential (equalization and normalization) ► polynomial SVM applied to pixels ► training set: about 2,500 face images (58x58 pixels)about 2,500 face images (58x58 pixels) about 10,000 non face images (extended to 13,000)about 10,000 non face images (extended to 13,000)

SVM: component-based detector (Poggio’s group) ► some preprocessing essential (equalization and normalization) ► two level system (always linear SVMs): component classifiers (14: eyes, nose,…)component classifiers (14: eyes, nose,…) geometrical configuration classifier based on maximal outputsgeometrical configuration classifier based on maximal outputs

global vs component-based ► component-based performs better (more robust to pose variation and/or occlusion) ► global a little faster (though they are both pretty slow, too many patches have to be stored!)

naive bayes (Kanade’s group) ► multiple detectors for different views (size and orientation) ► for each view: statistical modeling using predefined attribute histograms (17), about 2,000 face examples independency is required…independency is required… very good for out-of-plane rotation but involved procedure for building histograms (bootstrap, AdaBoost…)

AdaBoost (Viola & Jones) ► wavelet like features (computed efficiently) ► feature selected through AdaBoost (each weak classifier depends on a single feature) ► detection is obtained by training a cascade of classifiers ► very fast and effective on frontal faces

summing up ► SVM: components based on prior knowledge, simple, very good results but rather slow (optimization approaches…) ► naive bayes: robust against rotation, prior knowledge on feature selection, rather hand crafted statistical analysis, many models need to be learned (each with many examples) ► AdaBoost: data-driven feature selection, fast, frontal face only

what we do ► we assume we are given a fair number of positive examples only (no negatives) ► we want to explore how far one can get by combining fully data driven techniques based on 1D data ► we look at old-fashioned hypothesis testing (false negative rate under full control)

one possible way to object detection  building models can be expensive (domain dependent)  learning from positive examples only is more difficult, but…  classical hypothesis testing controls the false negative rate naturally

testing hypotheses ► HT with only one observation ► testing for independence with rank test (seems appropriate for comparing different features)

CBCL database faces (19x19pixels) training: 2429 test: 472 nonfaces (19x19pixels) training: 4548 test: 23573

training by hypothesis testing I.we first compute a large number of features (for the moment about 16,000) on the training set images II.a subset of good features (about 1,000) is then selected III.of these, a subset of independent features is considered (ending up with about 100) IV.multiple statistical tests are then constructed using the training set (one test for each feature)

image measurements ► grey value at fixed locations (19 x 19) ► tomographies (19 vertical + 19 horizontal + 37 along the 45deg diagonals) ► ranklets (5184 vertical, 5184 horizontal, 5184 diagonal) ► a total of about 16,000 features

ranklets (Smeraldi, 2002)

vertical ranklets (variance-to-natural support ratio)

a salient and a non-salient feature we discard all features for which the ratio falls below the threshold  0.15 (this leaves us with about 2000 features)

independent feature selection 1. we run independence tests on all possible pairs of salient features of the same category 2. we build a complete graph for each category with as many nodes as features. An edge between two features is deleted if for the two features the Spearman’s test rejects the independence hypothesis with probability 

independent feature selection 3. we then search the graph for maximally complete subgraphs (cliques) which we regard as sets of independent features for   0.5 we are left with 44 vertical, 64 horizontal, 35 diagonal ranklets, and 38 tomographies

testing for all image locations, all applicable scales and a fixed number  I.compute the values of the good, independent features II.perform multiple statistical tests at a certain confidence level III.a positive example is located if  tests are passed

multiple tests  we run multiple tests living with the fact that we won’t detect a certain fraction of the objects we want to find  luckily we are in a position to decide the fraction beforehand  we gain power because each test looks at a new feature

some results (franceschi et al, 2004) 472 positive vs 23,573 negatives tomographies + ranklets randomly chosen overlapping features

once you have detected a face… ask Thomas