A Unified Framework for Context Assisted Face Clustering

Slides:



Advertisements
Similar presentations
A Comparison of Implicit and Explicit Links for Web Page Classification Dou Shen 1 Jian-Tao Sun 2 Qiang Yang 1 Zheng Chen 2 1 Department of Computer Science.
Advertisements

Attributes for Classifier Feedback Amar Parkash and Devi Parikh.
Pseudo-Relevance Feedback For Multimedia Retrieval By Rong Yan, Alexander G. and Rong Jin Mwangi S. Kariuki
Recognizing Human Actions by Attributes CVPR2011 Jingen Liu, Benjamin Kuipers, Silvio Savarese Dept. of Electrical Engineering and Computer Science University.
Efficient summarization framework for multi-attribute uncertain data Jie Xu, Dmitri V. Kalashnikov, Sharad Mehrotra 1.
Foreground Focus: Finding Meaningful Features in Unlabeled Images Yong Jae Lee and Kristen Grauman University of Texas at Austin.
Carolina Galleguillos, Brian McFee, Serge Belongie, Gert Lanckriet Computer Science and Engineering Department Electrical and Computer Engineering Department.
An Infant Facial Expression Recognition System Based on Moment Feature Extraction C. Y. Fang, H. W. Lin, S. W. Chen Department of Computer Science and.
Learning Visual Similarity Measures for Comparing Never Seen Objects Eric Nowak, Frédéric Jurie CVPR 2007.
Patch to the Future: Unsupervised Visual Prediction
Identifying Image Spam Authorship with a Variable Bin-width Histogram-based Projective Clustering Song Gao, Chengcui Zhang, Wei Bang Chen Department of.
Face Detection, Pose Estimation, and Landmark Localization in the Wild
Intelligent Systems Lab. Recognizing Human actions from Still Images with Latent Poses Authors: Weilong Yang, Yang Wang, and Greg Mori Simon Fraser University,
IT’S NOT POLITE TO POINT: DESCRIBING PEOPLE WITH UNCERTAIN ATTRIBUTES CVPR 2013 Poster.
Liyan Zhang, Ronen Vaisenberg, Sharad Mehrotra, Dmitri V. Kalashnikov Department of Computer Science University of California, Irvine This material is.
Daozheng Chen 1, Mustafa Bilgic 2, Lise Getoor 1, David Jacobs 1, Lilyana Mihalkova 1, Tom Yeh 1 1 Department of Computer Science, University of Maryland,
1 Learning to Detect Objects in Images via a Sparse, Part-Based Representation S. Agarwal, A. Awan and D. Roth IEEE Transactions on Pattern Analysis and.
A Study of Approaches for Object Recognition
Local Affine Feature Tracking in Films/Sitcoms Chunhui Gu CS Final Presentation Dec. 13, 2006.
1 Integrating User Feedback Log into Relevance Feedback by Coupled SVM for Content-Based Image Retrieval 9-April, 2005 Steven C. H. Hoi *, Michael R. Lyu.
A Self-Organizing Approach to Background Subtraction for Visual Surveillance Applications Lucia Maddalena and Alfredo Petrosino, Senior Member, IEEE.
Smart Traveller with Visual Translator for OCR and Face Recognition LYU0203 FYP.
Face Processing System Presented by: Harvest Jang Group meeting Fall 2002.
Jacinto C. Nascimento, Member, IEEE, and Jorge S. Marques
Ordinal Decision Trees Qinghua Hu Harbin Institute of Technology
EVENT IDENTIFICATION IN SOCIAL MEDIA Hila Becker, Luis Gravano Mor Naaman Columbia University Rutgers University.
Learning Table Extraction from Examples Ashwin Tengli, Yiming Yang and Nian Li Ma School of Computer Science Carnegie Mellon University Coling 04.
Speaker: Chi-Yu Hsu Advisor: Prof. Jian-Jung Ding Leveraging Stereopsis for Saliency Analysis, CVPR 2012.
Wang, Z., et al. Presented by: Kayla Henneman October 27, 2014 WHO IS HERE: LOCATION AWARE FACE RECOGNITION.
Gender and 3D Facial Symmetry: What’s the Relationship ? Xia BAIQIANG (University Lille1/LIFL) Boulbaba Ben Amor (TELECOM Lille1/LIFL) Hassen Drira (TELECOM.
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.
Exploiting Ontologies for Automatic Image Annotation M. Srikanth, J. Varner, M. Bowden, D. Moldovan Language Computer Corporation
1 A Graph-Theoretic Approach to Webpage Segmentation Deepayan Chakrabarti Ravi Kumar
Dynamic 3D Scene Analysis from a Moving Vehicle Young Ki Baik (CV Lab.) (Wed)
Presenter: Lung-Hao Lee ( 李龍豪 ) January 7, 309.
Developing Trust Networks based on User Tagging Information for Recommendation Making Touhid Bhuiyan et al. WISE May 2012 SNU IDB Lab. Hyunwoo Kim.
Exploiting Context Analysis for Combining Multiple Entity Resolution Systems -Ramu Bandaru Zhaoqi Chen Dmitri V.kalashnikov Sharad Mehrotra.
Chao-Yeh Chen and Kristen Grauman University of Texas at Austin Efficient Activity Detection with Max- Subgraph Search.
Face Detection Ying Wu Electrical and Computer Engineering Northwestern University, Evanston, IL
Challenges in Mining Large Image Datasets Jelena Tešić, B.S. Manjunath University of California, Santa Barbara
A New Method for Automatic Clothing Tagging Utilizing Image-Click-Ads Introduction Conclusion Can We Do Better to Reduce Workload?
Creating Subjective and Objective Sentence Classifier from Unannotated Texts Janyce Wiebe and Ellen Riloff Department of Computer Science University of.
Data Mining, ICDM '08. Eighth IEEE International Conference on Duy-Dinh Le National Institute of Informatics Hitotsubashi, Chiyoda-ku Tokyo,
Face Image-Based Gender Recognition Using Complex-Valued Neural Network Instructor :Dr. Dong-Chul Kim Indrani Gorripati.
GENDER AND AGE RECOGNITION FOR VIDEO ANALYTICS SOLUTION PRESENTED BY: SUBHASH REDDY JOLAPURAM.
Unsupervised Auxiliary Visual Words Discovery for Large-Scale Image Object Retrieval Yin-Hsi Kuo1,2, Hsuan-Tien Lin 1, Wen-Huang Cheng 2, Yi-Hsuan Yang.
From Words to Senses: A Case Study of Subjectivity Recognition Author: Fangzhong Su & Katja Markert (University of Leeds, UK) Source: COLING 2008 Reporter:
Learning Photographic Global Tonal Adjustment with a Database of Input / Output Image Pairs.
Instance Discovery and Schema Matching With Applications to Biological Deep Web Data Integration Tantan Liu, Fan Wang, Gagan Agrawal {liut, wangfa,
Max-Confidence Boosting With Uncertainty for Visual tracking WEN GUO, LIANGLIANG CAO, TONY X. HAN, SHUICHENG YAN AND CHANGSHENG XU IEEE TRANSACTIONS ON.
Richer Human-Machine Communication in Attributes-based Visual Recognition Devi Parikh TTIC.
Motion Segmentation at Any Speed Shrinivas J. Pundlik Department of Electrical and Computer Engineering, Clemson University, Clemson, SC.
Shadow Detection in Remotely Sensed Images Based on Self-Adaptive Feature Selection Jiahang Liu, Tao Fang, and Deren Li IEEE TRANSACTIONS ON GEOSCIENCE.
Facial Smile Detection Based on Deep Learning Features Authors: Kaihao Zhang, Yongzhen Huang, Hong Wu and Liang Wang Center for Research on Intelligent.
Experience Report: System Log Analysis for Anomaly Detection
CNN-RNN: A Unified Framework for Multi-label Image Classification
Neighborhood - based Tag Prediction
Deeply learned face representations are sparse, selective, and robust
Guillaume-Alexandre Bilodeau
Detecting Sexually Provocative Images
Cheng-Ming Huang, Wen-Hung Liao Department of Computer Science
Attributes and Simile Classifiers for Face Verification
A weight-incorporated similarity-based clustering ensemble method based on swarm intelligence Yue Ming NJIT#:
Clustering Algorithms for Noun Phrase Coreference Resolution
Disambiguation Algorithm for People Search on the Web
Estimation of Skin Color Range Using Achromatic Features
Automatic Segmentation of Data Sequences
AHED Automatic Human Emotion Detection
AHED Automatic Human Emotion Detection
Presentation transcript:

A Unified Framework for Context Assisted Face Clustering Liyan Zhang, Dmitri V. Kalashnikov, Sharad Mehrotra Department of Computer Science University of California, Irvine

Introduction User Feedback Face Clustering Face Tagging Human is Center Explosion of Media Data User Feedback Face Clustering Face Tagging The prevalence of digital cameras as well as the emergence of online media web sites such as Flickr, Picasa, Fackbook, Twitter, etc., makes the creation, storage and sharing of multimedia content much easier than before, which leads to the explosion of massive media data. As the continue growing of the size of personal media collections, the problem of media organization, management and retrieval has become a much more pressing issue. Among most photo collections, human is usually the focus of images. To better understand and manage these

Outline Introduction to Face Clustering Traditional Approaches for Face Clustering The Proposed Context Assisted Framework Experimental Results Conclusions and Future Work

Face Appearance based Approach Facial Features Detected Faces …… Clustering Results Clustering Algorithm Face Similarity Graph

Appearance based Face Clustering Results Good Clustering Results High Precision,High Recall Tight Clustering Threshold High Precision, Low Recall loose Clustering Threshold Low Precision, High Recall Too Much Merging Work!

Drawbacks of Facial Similarities Same People Look Different Different Pose Different Expression Different Illumination Different Occlusion Different People Look The Same Boy Girl

Context Information Helps Common Scene: Geo Location Captured Time Image Background Social Context: People Co-occur Human Attributes: Age Ethnicity Gender Hair … Clothing: Cloth color

Related Work People Co-occurrence [1] Human Attributes [3] Clothing [2] Context Prior work Single Context Heterogeneous Face Level Cluster Level Single Context Type [1] Y. J. Lee and K. Grauman. Face discovery with social context. In BMVC, 2011. [3] N. Kumar and et al. Describable visual attributes for face verification and image search. In IEEE TPAMI, 2011. [2] A. Gallagher and T. Chen. Clothing cosegmentation for recognizing people. In IEEE CVPR, 2008. Heterogeneous Context Feature

The Framework …… …… …… … cont Initial Clusters : High Precision, Low Recall Photo Collection Detected Faces cont Common Scene People Co-occurrence Human Attributes Clothing … …… Iterative Merging …… Final Clusters: High Precision, High Recall

Context Features Extraction Same? Cluster level Common Scene People Co-occurrence Human Attributes Clothing Context Similarities Context Constraints Integrate Diff ?

Common Scene Image captured time, camera model, image visual features Same people I1 I2 I3 C2 C1 I1 I2

People Co-occurrence diff same

People Co-occurrence I1 I2 f 5 f 2 f 1 f 4 f 3 f 8 f 7 f 6 1 1 1 1 1 1 Cluster Co-Occurrence Graph 1 1 1 1 1 1

Human Attributes same diff 73-D 73-D N. Kumar and et al. Describable visual attributes for face verification and image search. In IEEE TPAMI, 2011.

Learn Weights From Dataset! Human Attributes Attribute C5 Only One Child Many Children! AGE attribute C5 cosine Similar? f 1 f 2 f 3 Attribute Different Attributes Different Weights Bootstrapping: Learn Weights From Dataset!

Human Attributes diff Face Attributes Label C1 ~ C1 Train Classifier

Clothing Time Sensitive! Diff Same Similarity from clothes Cloth color hist similarity Time diff Time slot threshold Time diff << S Time diff >> S

Cluster-Level Context Similarities Context Features Common Scene People Co-occurrence Cluster-Level Context Similarities Human Attributes Clothing

Cluster-Level Context Constraints Context Features Time & Location Time: t s Indoor Time: (t+1)s Outdoor Diff Diff Distinct Attributes Cluster-Level Context Constraints Diff Co-occurred people

Single Context Feature Fails People Co-occurrence Common Scene Same Same Diff Diff Co-occurred People Different Attributes Different Clothing

Integration is Required Context Similarities Context Constraints Integrate ? Aggregation? Set a rule? Context Features Y=a +b +c +d +e +… The importance of features differ with different dataset! Merge Not Learn Rules from Each Dataset!

How to Learn Rules? …… … Manually Label Learning Rules Apply Rules Training Dataset Context Constraints Diff Pairs Apply rules Learning Rules Training Dataset Automatic Label Split Initial clusters Same Pairs Bootstrapping …… Initial Clusters : High Precision, Low Recall Facial Features Photo Collection Learn from data Itself! cont Common Scene People Co-occurrence Human Attributes Clothing … Context Similarity & Constraint

Example of Automatic Labeling Splitting Training pairs Label Same Diff Diff same diff Cost-sensitive DTC

1st Splitting—Training--Predicting pairs Label Same Diff Diff Predicting pairs predict Same Diff … ( ): 5 same Cost-sensitive DTC ( ): 1 same

2nd Splitting—Training--Predicting pairs Label Same Diff Diff Predicting pairs predict Same Diff … ( ): 4 same Cost-sensitive DTC ( ): 0 same

Combine results pairs predict … 1st Time C1-C3: 5 same C2-C3: 1 same Diff … 1st Time C1-C3: 5 same C2-C3: 1 same C1-C3: 9 same pairs predict Same Diff … Merge C1-C3 C2-C3: 1 same 2nd Time C1-C3: 4 same C2-C3: 0 same

Unified Framework … Faces Photo Album splitting training prediction Extracted Faces Context Features Iterative Merging splitting training prediction Facial Pure clusters splitting training prediction Final Decision … splitting training prediction YES Merge Pairs? No Results

Experiment Datasets Surveillance Gallagher Wedding

Evaluation Metrics B-cubed Precision and Recall

Performance Comparison Photo Album Context Features Splitting Training Predicting Process Our Approach: Merge Decision Facial Features Pure Clusters Update Precision Recall Cluster Threshold 5095 Different Clusters Picasa: Context Similarities Aggregation Facial Similarities Different Parameter: p Different Clusters Affinity Propagation:

Results

Results High Precision Higher Recall

Results High Precision, 662 clusters 31 Real Person, 631 Merging 4 Times High Precision, 203 clusters 31 Real Person, 172 Merging

Results Less Clusters Less Manual Merging

Results

Conclusion and Future Work Single Context Feature Similarity Prior work Efficiency? User Feedback? Break points for precision dropping? Future work Our Approach Context Similarity Common Scene People Co-occur Human Attributes Clothing Bootstrapping Integration Iterative Merging High precision High recall Heterogeneous Context Features Context Constraint Co-occur People Distinct Attributes Time & Space

Thank you! Questions?