Visual Recognition of American Sign Language Using Hidden Markov Models 문현구 2019-05-24 문현구.

Slides:



Advertisements
Similar presentations
1 Gesture recognition Using HMMs and size functions.
Advertisements

We consider situations in which the object is unknown the only way of doing pose estimation is then building a map between image measurements (features)
Vogler and Metaxas University of Toronto Computer Science CSC 2528: Handshapes and Movements: Multiple- channel ASL recognition Christian Vogler and Dimitris.
Building an ASR using HTK CS4706
Character Recognition using Hidden Markov Models Anthony DiPirro Ji Mei Sponsor:Prof. William Sverdlik.
Large Vocabulary Unconstrained Handwriting Recognition J Subrahmonia Pen Technologies IBM T J Watson Research Center.
Yasuhiro Fujiwara (NTT Cyber Space Labs)
Speech Recognition Part 3 Back end processing. Speech recognition simplified block diagram Speech Capture Speech Capture Feature Extraction Feature Extraction.
Angelo Dalli Department of Intelligent Computing Systems
Designing Facial Animation For Speaking Persian Language Hadi Rahimzadeh June 2005.
Introduction to Hidden Markov Models
Hidden Markov Models Bonnie Dorr Christof Monz CMSC 723: Introduction to Computational Linguistics Lecture 5 October 6, 2004.
Hidden Markov Models Theory By Johan Walters (SR 2003)
Hidden Markov Models in NLP
HMM-BASED PATTERN DETECTION. Outline  Markov Process  Hidden Markov Models Elements Basic Problems Evaluation Optimization Training Implementation 2-D.
Rodent Behavior Analysis Tom Henderson Vision Based Behavior Analysis Universitaet Karlsruhe (TH) 12 November /9.
Face Recognition Using Embedded Hidden Markov Model.
Part 6 HMM in Practice CSE717, SPRING 2008 CUBS, Univ at Buffalo.
Evaluating the Quality of Image Synthesis and Analysis Techniques Matthew O. Ward Computer Science Department Worcester Polytechnic Institute.
Effective Gaussian mixture learning for video background subtraction Dar-Shyang Lee, Member, IEEE.
Learning Hidden Markov Model Structure for Information Extraction Kristie Seymour, Andrew McCullum, & Ronald Rosenfeld.
Learning and Recognizing Activities in Streams of Video Dinesh Govindaraju.
Handwritten Character Recognition using Hidden Markov Models Quantifying the marginal benefit of exploiting correlations between adjacent characters and.
Statistical Natural Language Processing. What is NLP?  Natural Language Processing (NLP), or Computational Linguistics, is concerned with theoretical.
May 20, 2006SRIV2006, Toulouse, France1 Acoustic Modeling of Accented English Speech for Large-Vocabulary Speech Recognition ATR Spoken Language Communication.
ETHEM ALPAYDIN © The MIT Press, Lecture Slides for.
Online Chinese Character Handwriting Recognition for Linux
Isolated-Word Speech Recognition Using Hidden Markov Models
1 7-Speech Recognition (Cont’d) HMM Calculating Approaches Neural Components Three Basic HMM Problems Viterbi Algorithm State Duration Modeling Training.
Feature and object tracking algorithms for video tracking Student: Oren Shevach Instructor: Arie nakhmani.
Knowledge Systems Lab JN 9/10/2002 Computer Vision: Gesture Recognition from Images Joshua R. New Knowledge Systems Laboratory Jacksonville State University.
Alignment and classification of time series gene expression in clinical studies Tien-ho Lin, Naftali Kaminski and Ziv Bar-Joseph.
Abstract Developing sign language applications for deaf people is extremely important, since it is difficult to communicate with people that are unfamiliar.
7-Speech Recognition Speech Recognition Concepts
General Tensor Discriminant Analysis and Gabor Features for Gait Recognition by D. Tao, X. Li, and J. Maybank, TPAMI 2007 Presented by Iulian Pruteanu.
International Conference on Intelligent and Advanced Systems 2007 Chee-Ming Ting Sh-Hussain Salleh Tian-Swee Tan A. K. Ariff. Jain-De,Lee.
Presented by: Fang-Hui Chu Boosting HMM acoustic models in large vocabulary speech recognition Carsten Meyer, Hauke Schramm Philips Research Laboratories,
22CS 338: Graphical User Interfaces. Dario Salvucci, Drexel University. Lecture 10: Advanced Input.
Experimental Results Abstract Fingerspelling is widely used for education and communication among signers. We propose a new static fingerspelling recognition.
A NOVEL METHOD FOR COLOR FACE RECOGNITION USING KNN CLASSIFIER
Context-based vision system for place and object recognition Antonio Torralba Kevin Murphy Bill Freeman Mark Rubin Presented by David Lee Some slides borrowed.
Performance Comparison of Speaker and Emotion Recognition
MINIMUM WORD CLASSIFICATION ERROR TRAINING OF HMMS FOR AUTOMATIC SPEECH RECOGNITION Yueng-Tien, Lo Speech Lab, CSIE National.
GENDER AND AGE RECOGNITION FOR VIDEO ANALYTICS SOLUTION PRESENTED BY: SUBHASH REDDY JOLAPURAM.
Discriminative Training and Machine Learning Approaches Machine Learning Lab, Dept. of CSIE, NCKU Chih-Pin Liao.
Statistical Models for Automatic Speech Recognition Lukáš Burget.
Phone-Level Pronunciation Scoring and Assessment for Interactive Language Learning Speech Communication, 2000 Authors: S. M. Witt, S. J. Young Presenter:
Definition of the Hidden Markov Model A Seminar Speech Recognition presentation A Seminar Speech Recognition presentation October 24 th 2002 Pieter Bas.
1 7-Speech Recognition Speech Recognition Concepts Speech Recognition Approaches Recognition Theories Bayse Rule Simple Language Model P(A|W) Network Types.
Student Gesture Recognition System in Classroom 2.0 Chiung-Yao Fang, Min-Han Kuo, Greg-C Lee, and Sei-Wang Chen Department of Computer Science and Information.
A NONPARAMETRIC BAYESIAN APPROACH FOR
Hidden Markov Models BMI/CS 576
Learning, Uncertainty, and Information: Learning Parameters
Online Multiscale Dynamic Topic Models
EEL 6586: AUTOMATIC SPEECH PROCESSING Hidden Markov Model Lecture
Supervised Time Series Pattern Discovery through Local Importance
Posture Monitoring System for Context Awareness in Mobile Computing
Statistical Models for Automatic Speech Recognition
Pattern Recognition Sergios Theodoridis Konstantinos Koutroumbas
An INTRODUCTION TO HIDDEN MARKOV MODEL
Hidden Markov Models - Training
Computational NeuroEngineering Lab
Statistical Models for Automatic Speech Recognition
Three classic HMM problems
4.0 More about Hidden Markov Models
Handwritten Characters Recognition Based on an HMM Model
LECTURE 15: REESTIMATION, EM AND MIXTURES
Morphological Segmentation of Natural Gesture
Introduction to HMM (cont)
Sign Language Recognition With Unsupervised Feature Learning
Presentation transcript:

Visual Recognition of American Sign Language Using Hidden Markov Models 문현구 2019-05-24 문현구

Introduction(1) Like speech recognition and handwriting recognition, visual recognition is ideal domain of HMM To date, most work on sign language recognition has employed expensive wired “datagloves” In this paper, it requires that the user wear inexpensive colored gloves 2019-05-24 문현구

Introduction(2) The goal of this work is to allow the creation of a real-time system by guaranteeing each separate component (tracking, analysis, and recognition) runs in real-time 2019-05-24 문현구

Machine Sign Language Recognition Former systems have generally concentrated on isolated signs and small training and test sets 2019-05-24 문현구

Hidden Markov Modeling(1) A four state HMM with skip transitions was determined to be sufficient for the initial topology 2019-05-24 문현구

Hidden Markov Modeling(2) Three key problems in HMM use The evaluation problem : The estimation problem : The decoding problem : Optimal state sequence 2019-05-24 문현구

Hidden Markov Modeling(3) The evaluation problem 2019-05-24 문현구

Hidden Markov Modeling(4) The estimation problem 2019-05-24 문현구

Hidden Markov Modeling(5) Modification to handle a variety of characteristic densities: 2019-05-24 문현구

Hidden Markov Modeling(6) The decoding problem 2019-05-24 문현구

Recovering Hands from Video The subject wears distinctly colored gloves on each hand(a yellow for the right hand and an orange for the left) To find each hand initially, the algorithm scans the image until it finds a pixel of the appropriate color. 2019-05-24 문현구

Feature Extraction An eight element feature vector consisting of each hand’s x and y position, angle of axis of least inertia, and eccentricity of bounding ellipse was chosen 2019-05-24 문현구

Experimentation(1) 494 sentences were collected, each sign ranged from approximately 1 to 3 seconds in length, and no intentional pauses were placed between signs within a sentence, but the sentences themselves were distinct 2019-05-24 문현구

Experimentation(2) Initial estimates for the means and variances of the output probabilities were provided by iteratively using Viterbi alignment on training data. Recomputing the means and variances by pooling the vectors in each segment. Entropic’s HMM ToolKit(HTK)is used as a basis for this step and all other HMM modeling and training tasks. 2019-05-24 문현구

Experimentation(3) The results from the initial alignment program are fed into a Baum-Welch re-estimator, whose estimates are, in turn, refined in embedded training which ignores any initial segmentation. For recognition, HTK’s Viterbi recognizer is used both with and without a strong grammar based on the known form of the sentences. 2019-05-24 문현구

Experimentation(4) The accuracy measure is calculated by subtracting the number of insertion errors from the number of correct labels and dividing by the total number of signs. 2019-05-24 문현구

Analysis and Discussion(1) The high recognition rate on the training data indicates that the HMM topologies are sound and that the models are converging. The 0.8% error rate on the independent test set shows that the models are generalizing well 2019-05-24 문현구

Analysis and Discussion(2) Some basic improvements Use deltas instead of absolute positions Add finger and palm tracking information Collect appropriate domain or task oriented data perform context modeling 2019-05-24 문현구

Conclusion Through use of hidden Markov models low error rates were achieved on both the training set and an independent test set without invoking complex models of the hands. With a larger training set and context modeling, lower error rates are expected. 2019-05-24 문현구