Face Detection Final Presentation Mark Lee Nic Phillips Paul Sowden Andy Tait 9 th May 2006.

Slides:



Advertisements
Similar presentations
A Graph based Geometric Approach to Contour Extraction from Noisy Binary Images Amal Dev Parakkat, Jiju Peethambaran, Philumon Joseph and Ramanathan Muthuganapathy.
Advertisements

Road-Sign Detection and Recognition Based on Support Vector Machines Saturnino, Sergio et al. Yunjia Man ECG 782 Dr. Brendan.
Evaluating Color Descriptors for Object and Scene Recognition Koen E.A. van de Sande, Student Member, IEEE, Theo Gevers, Member, IEEE, and Cees G.M. Snoek,
Face Detection & Synthesis using 3D Models & OpenCV Learning Bit by Bit Don Miller ITP, Spring 2010.
Electrical & Computer Engineering Dept. University of Patras, Patras, Greece Evangelos Skodras Nikolaos Fakotakis.
Simple Face Detection system Ali Arab Sharif university of tech. Fall 2012.
Image Processing IB Paper 8 – Part A Ognjen Arandjelović Ognjen Arandjelović
Performance Evaluation Measures for Face Detection Algorithms Prag Sharma, Richard B. Reilly DSP Research Group, Department of Electronic and Electrical.
Facial feature localization Presented by: Harvest Jang Spring 2002.
Color Image Processing
A Review of “Adaptive fingerprint image enhancement with fingerprint image quality analysis”, by Yun & Cho Malcolm McMillan.
ICIP 2000, Vancouver, Canada IVML, ECE, NTUA Face Detection: Is it only for Face Recognition?  A few years earlier  Face Detection Face Recognition 
Zhengya Xu, Hong Ren Wu, Xinghuo Yu, Fellow, IEEE, Bin Qiu, Senior Member, IEEE Colour Image Enhancement by Virtual Histogram Approach IEEE Transactions.
EE 7730 Image Segmentation.
Robust Object Segmentation Using Adaptive Thresholding Xiaxi Huang and Nikolaos V. Boulgouris International Conference on Image Processing 2007.
Face Detection: a Survey Speaker: Mine-Quan Jing National Chiao Tung University.
UNIVERSITY OF MURCIA (SPAIN) ARTIFICIAL PERCEPTION AND PATTERN RECOGNITION GROUP REFINING FACE TRACKING WITH INTEGRAL PROJECTIONS Ginés García Mateos Dept.
Facial Features Extraction Amit Pillay Ravi Mattani Amit Pillay Ravi Mattani.
Smart Traveller with Visual Translator. What is Smart Traveller? Mobile Device which is convenience for a traveller to carry Mobile Device which is convenience.
Triangle-based approach to the detection of human face March 2001 PATTERN RECOGNITION Speaker Jing. AIP Lab.
A Novel 2D To 3D Image Technique Based On Object- Oriented Conversion.
California Car License Plate Recognition System ZhengHui Hu Advisor: Dr. Kang.
Face Processing System Presented by: Harvest Jang Group meeting Fall 2002.
Colour Image Compression by Grey to Colour Conversion Mark S. Drew 1, Graham D. Finlayson 2, and Abhilash Jindal 3 1 Simon Fraser University 2 University.
Real-Time Vision on a Mobile Robot Platform Mohan Sridharan Joint work with Peter Stone The University of Texas at Austin
VINCENT URIAS, CURTIS HASH Detection of Humans in Images Using Skin-tone Analysis and Face Detection.
FACE DETECTION AND RECOGNITION By: Paranjith Singh Lohiya Ravi Babu Lavu.
Color Space for Skin Detection – A Review
Proceedings of the 2008 IEEE International Conference on Robotics and Biomimetics Bangkok, Thailand, February , 2009 Prefessor : 謝銘原 Student : 謝琮閔.
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.
Automated Face Detection Peter Brende David Black-Schaffer Veni Bourakov.
Vehicle License Plate Detection Algorithm Based on Statistical Characteristics in HSI Color Model Instructor : Dr. K. R. Rao Presented by: Prasanna Venkatesh.
Facial Feature Extraction Yuri Vanzine C490/B657 Computer Vision.
Joon Hyung Shim, Jinkyu Yang, and Inseong Kim
1 Webcam Mouse Using Face and Eye Tracking in Various Illumination Environments Yuan-Pin Lin et al. Proceedings of the 2005 IEEE Y.S. Lee.
Object Detection with Discriminatively Trained Part Based Models
ECE738 Advanced Image Processing Face Detection IEEE Trans. PAMI, July 1997.
1 Research Question  Can a vision-based mobile robot  with limited computation and memory,  and rapidly varying camera positions,  operate autonomously.
Creating Better Thumbnails Chris Waclawik. Project Motivation Thumbnails used to quickly select a specific a specific image from a set (when lacking appropriate.
` Tracking the Eyes using a Webcam Presented by: Kwesi Ackon Kwesi Ackon Supervisor: Mr. J. Connan.
Autonomous Robots Vision © Manfred Huber 2014.
By Naveen kumar Badam. Contents INTRODUCTION ARCHITECTURE OF THE PROPOSED MODEL MODULES INVOLVED IN THE MODEL FUTURE WORKS CONCLUSION.
Face Detection Using Skin Color and Gabor Wavelet Representation Information and Communication Theory Group Faculty of Information Technology and System.
Video Camera Security and Surveillance System ICAMES 2008 Team Members : Semih Altınsoy Denis Kürov Team Advisor: Assist. Prof. M. Elif Karslıgil May,
By: David Gelbendorf, Hila Ben-Moshe Supervisor : Alon Zvirin
A generic face processing framework: Technologies, Analyses and Applications Supervised by: Prof. Michael R. Lyu Presented by Jang Kim Fung Oral Defense.
Introduction Performance of metric learning is heavily dependent on features extracted Sensitive to Performance of Filters used Need to be robust to changes.
Face Detection Using Neural Network By Kamaljeet Verma ( ) Akshay Ukey ( )
Color Image Segmentation Mentor : Dr. Rajeev Srivastava Students: Achit Kumar Ojha Aseem Kumar Akshay Tyagi.
May 2003 SUT Color image segmentation – an innovative approach Amin Fazel May 2003 Sharif University of Technology Course Presentation base on a paper.
Vision Based hand tracking for Interaction The 7th International Conference on Applications and Principles of Information Science (APIS2008) Dept. of Visual.
Face Detection 蔡宇軒.
Content Based Coding of Face Images
EE368 Final Project Spring 2003
COMP 9517 Computer Vision Binary Image Analysis 4/15/2018
Introduction to Skin and Face Detection
Author : Sang Hwa Lee, Junyeong Choi, and Jong-Il Park
Color Image Processing
Face Detection EE368 Final Project Group 14 Ping Hsin Lee
CSSE463: Image Recognition Day 21
Traffic Sign Recognition Using Discriminative Local Features Andrzej Ruta, Yongmin Li, Xiaohui Liu School of Information Systems, Computing and Mathematics.
Color Image Processing
Text Detection in Images and Video
Statistical Approach to a Color-based Face Detection Algorithm
CSSE463: Image Recognition Day 23
Color Image Processing
De- rotating face - Procedure Eye detection and mouth detection
CSSE463: Image Recognition Day 23
Speaker: YI-JIA HUANG Date: 2011/12/08 Authors: C. N
ECE738 Final Project Face Detection Baseline
Presentation transcript:

Face Detection Final Presentation Mark Lee Nic Phillips Paul Sowden Andy Tait 9 th May 2006

Overview Detect multiple faces in images and return their locations  Pre-processor to face recognition  People tracking  Expression analysis Challenges include:  Pose, occlusion/beards/glasses etc, expression, orientation, lighting conditions

Our Chosen Techniques Compare 2 methods:  Feature Based Selection of image processing methods  Machine Learning (AdaBoost) Currently the de facto method C++ implementation (with CImg library)

Feature Based Method Uses image processing techniques to extract low-level facial features 1. Identify and segment possible regions of interest  Skin tones and overall proportions of skin regions. 2. Determine if each region contains face  Locate facial features – eyes and mouth  Compare feature locations using rules-based approach

Skin Colour Map Many different implementations Chosen Hsu model – good performance and parameters available Non-linear mapping of YCrCb space Ellipse in CrCb plane defines skin colour Works for multiple skin types (from Hsu et al)

Colour Correction Choose ‘white’ reference level from image based on maximum 5% luminance values Find scale factor to make this pure white in RGB space Apply same scaling to all other image pixels Colour corrected imageImage with red tint

Colour Correction Benefits Original Image Colour Corrected Image

Region Finding Algorithm Two Pass:  Identify local region connectivity and build equivalence map  Re-label equivalences Remove Regions:  Small area (noise)  Min. dimensions  Aspect ratio Output:  Set of bounding boxes containing possible face candidates  Passed to Eye+Mouth feature detection stage Connectivity Region Neighbourhood Detected Face Candidates

Skin Adjacency Problem Overlapping regions of skin colour cause false detection Skin Map

Updated Connectivity Criteria New connectivity requirements:  Is a Skin pixel  Y ±8 between adj regions Still can be problematic with some images with inadequate Y variation Edge criteria tested to augment Y channel connectivity metric  Faces broken into separate regions by glasses, eyebrows, strong lighting  Decided that performance gain was not worth the complexity increase Y Channel

Localisation of facial features Using the skin tone bounding boxes, aim to located facial features. Eye and mouth detection based on colour information and morphological techniques. Using YC b C r colour space. Then use heuristic method to discriminate between Face and Not-Face

Eye Detection - Chrominance Chrominance eyemap:  High C b and low C r values around eyes  High response in C b /C r  Also divide by original luminance – eyes usually dark areas. OriginalChrominance eyemap

Eye Detection - Luminance Greyscale morphological methods to emphasise areas of high luminance change – e.g. eyes Results – highlights eyes, plus sometimes nose, open mouthes Combine with chrominance map to pinpoint eye regions. LuminanceDilated luminance LuminanceLuminance + Crominance

Mouthmap Mouth stronger red and weaker blue  C r higher than C b  Low response in C r /C b  High response in C r 2  η based on ratio of C r 2 to C r /C b Original Mouthmap

Results Mouth – high luminance High chrominance map on dark hair Hair – high chrominance, but low luminance Luminance affected by reflection Blue – based on luminance dilation peaks Red – chrominance eyemap/luminance peaks Green – mouthmap peak

Heuristic Rules Eliminate false eye candidates:  Merge repeated eye points  Eye radial distance from mouth  Distance above mouth  Eye-eye-mouth triangle – min and max angles Removal of false eye candidates

Feature Based Methods Conclusions Face detection correct in many cases. Skin detection and segmentation works well  Sometimes confused by overlapping regions in complex arrangements Eye and mouth detection mostly correct  Red-eye, jewellery, other skin areas with high contrast differences can cause false-positives Heuristic elimination of false positives works well and fast  Additional rules could improve robustness Speed improvements possible for the intensive segmentation stage. Subsequent stages very fast.

[1] S. Cooray and N. O’Connor. A Hybrid Technique for Face Detection in Color Images. In Proc. IEEE Conference on Advanced Video and Signal Based Surveillance, pages , [2] R. Hsu and M. Abdel-Mottaleb. Face Detection in Color Images. In IEEE Transactions on Pattern Analysis and Machine Intelligence, volume 25 issue 5, pages , 2002