Vineeth Balasubramanian Shayok Chakraborty Sreekar Krishna Sethuraman Panchanathan C ENTER FOR C OGNITIVE U BIQUITOUS C OMPUTING CUbiC Human Centered Machine.

Slides:



Advertisements
Similar presentations
Pseudo-Relevance Feedback For Multimedia Retrieval By Rong Yan, Alexander G. and Rong Jin Mwangi S. Kariuki
Advertisements

A Unified Framework for Context Assisted Face Clustering
EE462 MLCV Lecture 5-6 Object Detection – Boosting Tae-Kyun Kim.
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.
Digital Interactive Entertainment Dr. Yangsheng Wang Professor of Institute of Automation Chinese Academy of Sciences
Electrical & Computer Engineering Dept. University of Patras, Patras, Greece Evangelos Skodras Nikolaos Fakotakis.
Online Multiple Classifier Boosting for Object Tracking Tae-Kyun Kim 1 Thomas Woodley 1 Björn Stenger 2 Roberto Cipolla 1 1 Dept. of Engineering, University.
Joint Eye Tracking and Head Pose Estimation for Gaze Estimation
F ACE D ETECTION FOR A CCESS C ONTROL By Dmitri De Klerk Supervisor: James Connan.
EE462 MLCV Lecture 5-6 Object Detection – Boosting Tae-Kyun Kim.
São Paulo Advanced School of Computing (SP-ASC’10). São Paulo, Brazil, July 12-17, 2010 Looking at People Using Partial Least Squares William Robson Schwartz.
Real-Time Non-Rigid Shape Recovery via AAMs for Augmented Reality Jackie Zhu Oct. 24, 2006.
Rapid Object Detection using a Boosted Cascade of Simple Features
Game Development with Kinect
CS335 Principles of Multimedia Systems Multimedia and Human Computer Interfaces Hao Jiang Computer Science Department Boston College Nov. 20, 2007.
MACHINE VISION GROUP Multimodal sensing-based camera applications Miguel Bordallo 1, Jari Hannuksela 1, Olli Silvén 1 and Markku Vehviläinen 2 1 University.
Wen-Chyi Lin CS2310 Software Engineering.  “Never express yourself more clearly than you are able to think” by Niels Bohr. However, there are times and.
Wang, Z., et al. Presented by: Kayla Henneman October 27, 2014 WHO IS HERE: LOCATION AWARE FACE RECOGNITION.
- Talkback with Dark screen Rapid key input and Speak PW - Font Size - Negative Colors - Magnification gestures - Notification reminder - Colour adjustment.
Exploiting SenseCam for Helping the Blind in Business Negotiations Shuaib Karim, Amin Andjomshoaa, A Min Tjoa (skarim, andjo, Institute.
Multimedia Specification Design and Production 2013 / Semester 2 / week 8 Lecturer: Dr. Nikos Gazepidis
2003Lenko Grigorov, CISC 839 eyePROXY Lenko Grigorov, CISC 839 Supervisor: Roel Vertegaal Additional support by Skaburskis A and Changuk S.
A General Framework for Tracking Multiple People from a Moving Camera
A Method for Hand Gesture Recognition Jaya Shukla Department of Computer Science Shiv Nadar University Gautam Budh Nagar, India Ashutosh Dwivedi.
C ENTER FOR C OGNITIVE U BIQUITOUS C OMPUTING CUbiC ARIZONA STATE UNIVERSITY Sreekar Krishna, Vineeth Balasubramanian, Sethuraman (Panch) Panchanathan.
ACTIVE LEARNING USING CONFORMAL PREDICTORS: APPLICATION TO IMAGE CLASSIFICATION HypHyp Introduction HypHyp Conceptual overview HypHyp Experiments and results.
VibroGlove An Assistive Technology Aid for Conveying Facial Expressions Sreekar Krishna †, Shantanu Bala, Troy McDaniel, Stephen McGuire &Sethuraman Panchanathan.
DIEGO AGUIRRE COMPUTER VISION INTRODUCTION 1. QUESTION What is Computer Vision? 2.
Enhancing Human-Machine Communication via Visual Attributes Devi Parikh Virginia Tech.
Cerberus: A Context-Aware Security Scheme for Smart Spaces presented by L.X.Hung u-Security Research Group The First IEEE International Conference.
ENTERFACE 08 Project 1 “MultiParty Communication with a Tour Guide ECA” Mid-term presentation August 19th, 2008.
Designing for energy-efficient vision-based interactivity on mobile devices Miguel Bordallo Center for Machine Vision Research.
Face Detection Ying Wu Electrical and Computer Engineering Northwestern University, Evanston, IL
1 1 Spatialized Haptic Rendering: Providing Impact Position Information in 6DOF Haptic Simulations Using Vibrations 9/12/2008 Jean Sreng, Anatole Lécuyer,
Aiming Computing Technology at Enhancing the Quality of Life of People with ALS Some Sketches on Directions in Minimal Signaling Communication Communication.
Robust Nighttime Vehicle Detection by Tracking and Grouping Headlights Qi Zou, Haibin Ling, Siwei Luo, Yaping Huang, and Mei Tian.
IEEE 2015 Conference on Computer Vision and Pattern Recognition Active Learning for Structured Probabilistic Models with Histogram Approximation Qing SunAnkit.
Max-Confidence Boosting With Uncertainty for Visual tracking WEN GUO, LIANGLIANG CAO, TONY X. HAN, SHUICHENG YAN AND CHANGSHENG XU IEEE TRANSACTIONS ON.
C ENTER FOR C OGNITIVE U BIQUITOUS C OMPUTING CUbiC ARIZONA STATE UNIVERSITY Sreekar Krishna Committee: Dr. Sethuraman (Panch) Panchanathan, Chair Dr.
Under Guidance of Mr. A. S. Jalal Associate Professor Dept. of Computer Engineering and Applications GLA University, Mathura Presented by Dev Drume Agrawal.
Guillaume-Alexandre Bilodeau
AHED Automatic Human Emotion Detection
CALO VISUAL INTERFACE RESEARCH PROGRESS
Ubiquitous Computing and Augmented Realities
Semantic Video Classification
Foundations of Technology Information and Communication
Compositional Human Pose Regression
Introductory Seminar on Research: Fall 2017
Techshare Pro | BREAKTHROUGH WEARABLE ASSISTIVE TECHNOLOGY
Continuous Automated Chatbot Testing
Facial Recognition [Biometric]
Interactive Learning An empFinesseTM Smart Atomic Learning Solution.
Social neuroscience Domina Petric, MD.
CSc4730/6730 Scientific Visualization
But what are you telling me?
Wearable Visual Information Systems (VINST)
Learning complex visual concepts
Pilar Orero, Spain Yoshikazu SEKI, Japan 2018
Towards lifelike Computer Interfaces that learn
Human-centered Interfaces
3rd Studierstube Workshop TU Wien
Liyuan Li, Jerry Kah Eng Hoe, Xinguo Yu, Li Dong, and Xinqi Chu
Presented by: Mónica Domínguez
AHED Automatic Human Emotion Detection
Jie Chen, Shiguang Shan, Shengye Yan, Xilin Chen, Wen Gao
Human-object interaction
MAPVI: Meeting Accessibility for Persons with Visual Impairments
Computer Vision Readings
THE ASSISTIVE SYSTEM SHIFALI KUMAR BISHWO GURUNG JAMES CHOU
Presentation transcript:

Vineeth Balasubramanian Shayok Chakraborty Sreekar Krishna Sethuraman Panchanathan C ENTER FOR C OGNITIVE U BIQUITOUS C OMPUTING CUbiC Human Centered Machine Learning in a Social Interaction Assistant for Individuals with Visual Impairments ARIZONA STATE UNIVERSITY December 10, 2009

CUbiC Outline Introduction – Human Centered Machine LearningOverview of the Social Interaction AssistantExamples of Human Centered Machine LearningRelated Machine Learning Contributions

CUbiC Outline Introduction – Human Centered Machine LearningOverview of the Social Interaction AssistantExamples of Human Centered Machine LearningRelated Machine Learning Contributions

CUbiC Assistive Devices for the Visually Impaired Need to enrich the interaction of blind individuals with other individuals

CUbiC Social Interaction Hand Gestures Posture Eye Gaze Hair Clothing About 65% of the information during social interaction is communicated using non-verbal cues Krishna et al. A Systematic Requirements Analysis and Development of an Assistive Device to enhance the Social Interaction of people who are Blind or Visually Impaired. ECCV 2008

CUbiC Social Interaction for the blind The Need 1.1 million people in the US are legally blind, 37 million worldwide They face the fundamental challenges during social interaction Necessitates design of a Social Interaction Assistant for the visually impaired The Challenges Challenges such as person recognition/tracking, head pose estimation, gesture recognition, expression recognition to be addressed

CUbiC “The Human-Machine System” Human Machine  The human and the machine should be treated as a single entity  Both should jointly be used to deliver outputs

CUbiC Human-Centered Machine Learning Use the ability of the user to derive effective solutions

CUbiC Outline Introduction – Human Centered Machine LearningOverview of the Social Interaction AssistantExamples of Human Centered Machine LearningRelated Machine Learning Contributions

CUbiC The Social Interaction Assistant Haptic Belt Krishna et al. A wearable face recognition system for individuals with visual impairments. ACM SIG ASSETS, 2005 McDaniel et al. Using a Haptic Belt to convey Non Verbal Communication Cues during Social Interactions to Individuals who are Blind. IEEE HAVE, 2008

CUbiC Outline Introduction – Human Centered Machine LearningOverview of the Social Interaction AssistantExamples of Human Centered Machine LearningRelated Machine Learning Contributions

Integrated Face Localization/Recognition CUbiC Wearable camera on user Face detection (Viola – Jones: Adaboost) Face detected to my far right Face Recognition – an integral component in a Social Interaction Assistant

Integrated Face Localization/Recognition CUbiC Edwards et al. A Pragmatic Approach to the Design and Implementation of a Vibrotactile Belt and its Applications, IEEE HAVE, 2009 Direction conveyed through a vibrotactile cue in the haptic belt Human in the loop simplifies the problem The individual turns in the direction of vibration – camera captures frontal images

User Conformal Confidence Measures CUbiC  CP framework is used to quantify the level of certainty or “confidence” in decision making in machine learning applications.  Results are well-calibrated  Frequency of errors,, made by the system is exactly controlled by the confidence level, 1 -, defined by the user Shafer, Vovk. A Tutorial on Conformal Predictions, JMLR 2008

CUbiC Outline Introduction – Human Centered Machine LearningOverview of the Social Interaction AssistantExamples of Human Centered Machine LearningRelated Machine Learning Contributions

Online Active Learning for Person Recognition CUbiC Current Unlabeled Data Point Class A P-Value: p1 Class B P-Value: p2 Class C P-Value: p3 Largest eigenvalue used as a measure of discrepancy between the p-values V. Balasubramanian, S. Chakraborty and S. Panchanathan. Generalized Query by Transduction for Online Active Learning, IEEE ICCV 2009

Context Aware Batch Mode Active Learning CUbiC Data from a video stream  Select a batch of informative samples from a video stream to update the classifiers  Use of context based priors can further help select the salient instances  Select a batch of informative samples from a video stream to update the classifiers  Use of context based priors can further help select the salient instances Select the salient examples

Learning from Multiple Sources CUbiC Learning from Multiple Sources of information

CUbiC Questions ??.. Thank You !!..