Large-Scale Semantics for Image and Video Retrieval

Size: px
Start display at page:

Download "Large-Scale Semantics for Image and Video Retrieval"

Transcription

1 Large-Scale Semantics for Image and Video Retrieval Lexing Xie with Apostol Natsev, John Smith, Matthew Hill, John Kender, Quoc-Bao Nyugen, *Jelena Tesic, *Rong Yan, *Alex Haubold IBM T J Watson Research Center, Hawthorne, IQR workshop, Sept 20,

2 The Challenge of Large Scale The data view and the semantics view 2

3 Outline: Scaling Up Visual Semantics Multi-modal retrieval with visual semantics. [MM 05+07,CIVR ICME 06+07, TV 04-07] Crowd-sourced semantics in real-world events. Find snow capped mountains Construction and detection of multi-faceted visual taxonomy. [MM 10] The value and accuracy of image tags. Outdoors? Sky? Ship? Red? Grass? Flower? Water? [CIVR 10] Tagging vs. browsing for image annotation. [CVPR 08] 3

4 Retrieval Approaches for Multi-modal Search queries retrieval approaches Find shots of Condoleezza Rice Find shots of soccer game with goalpost visible Find scenes of snow capped mountains Text????? Image similarity ~????? 4 Semantic concepts ~???? Government Leader US Flag Face Sports Vegetation Walking/Running Mountain Snow Skiing Challenge: Leverage limited vocabulary of concepts to answer arbitrary queries? [Natsev et. al. MM 07] 4

5 Overview of Concept-Based Retrieval Methods Lexical Statistical Text Haubold et al. [TV 05-06, ICME 06] Kennedy et al. [TV 05] Neo et al. [TV 05,CIVR 06] Snoek et al. [TV 06] Ontology Map [MM 07] Term Co-occurrence Map [MM 07] Visual examples --- Smith et al. [ICME 03] Natsev et al. [KDD 04,MM 05] Rasiwasia et al. [CIVR 06] Snoek et al. [TV 06] Tešić et al [CIVR 07] Feature Selection [MM 07] Retrieval results --- Yan and Hauptmann [TV 06, CMU PhD 06] Kennedy and Chang [CIVR 07] 5 TV = TREC Video Retrieval (TRECVID) Workshop 5

6 Mapping Words to Concepts Lexical: synonym expansion Trigger concept if matching one of the (stemmed) synonyms Weight concepts using trigger frequency C: Airplane W: Airplane Plane Airline Flight Lexical: WordNet similarity Based on graph structure, statistics, etc. Lesk, Resnik, Jiang-Conrath, Wu.. Lexical: Ontology mapping Semantically tag queries and concepts [IBM PIQUANT] Incorporate correlations and logical relations Statistical: co-occurrence with significance Compute G^2 score for each word-concept pair Capture both positive and negative relations O i ln (O i / E i ) G 2 = 2 * Σ i word! word concept O 1 O 2! concept O 3 O 4 6

7 Mapping Images to Concepts Content-based feature selection Obtain concept distribution in query examples Compare to those in corpus background Visual query examples H1 H0 Query modeling with few examples View queries in different vector spaces: image feature + concept Build bagged SVM classifiers for retrieval Probabilistic local context analysis (plca) Select semantic concepts and learn weights using initial retrieval results Use top/bottom retrieved documents as pseudo positive/negative examples Initial Results Airplane N3 N3 N3 Sky O Road N1 N2 N1 P P N3 N1 P N2 N1 P P N3 P N2 N2 N2 N1 N1 N2 N2 N1 N1 N2 Y 1 Y m Video 1 X Video M w 1 :? w C :? 7

8 Results of Concept Mapping Query: Find shots of soccer game with goalpost visible. concept triggered? approaches WordNet ontology G^2 stats plca visual sports Y Y Y Y Y person Y walking_running Y Y Y Y Y vegetation Y Y Y Evaluate on TRECVID benchmark corpora Multi-lingual news from 6 channels 2005: 110 hours (77,816 shots) 2006: 160 hours (119,490 shots) Speech transcripts (ASR and MT) 39 LSCOM-Lite concept detectors 24 distinct multi-modal queries for 2005 and 2006 Measure AP (average precision) on each query Present improvement over baseline (in %) news videos ENG CHN ARB Average precision Precision AP Recall 8

9 Lexical Approaches: Evaluation on TRECVID 2005 Concept-based re-ranking is effective for retrieval Improvements vary for different queries Each lexical approach has different strength All approaches perform comparably overall ~20%+ 9 9

10 Multi-modal Approaches: Evaluation on TRECVID 2006 Text queries Multimodal queries Text-based expansion approaches perform comparably but are complementary Content-based approaches bring significant further improvements 10

11 The Need for a Visual Semantic Structure We need organization beyond just a few visual concepts For people to remember and keep track To generate *Negative* training examples To help improve prediction results meeting [Xie et. al. MM 10] male portrait A. Jolie party snow Isolated dots: Difficult to keep track of beyond just a few; Does not help create training data; Ignores inner semantic structure Concept Web: Take into account statistical structure in training+classification. Difficult to keep track of beyond just a few; Does not help create training data; Ignores inner semantic structure [Naphade 01,Yan 06, Jiang 07, Qi 07 ] Concept Tree: Hierarchical structure to help training. Difficult to author beyond just a few; rigid, relationship often violated, [Wu et al, Schimd et al, He+Zemel, ] 11

12 What Relations To Model? From word semantics Visually detectable concrete nouns + some verbs and adjectives Hypernyms and hyponyms: parent-children relationships and mutual exclusion NOT modeled: part-of, entailment, highly view-point and context dependent From image appearances a picture is a sentence -- JRK Multiple divisions for a concept building, urban, protest, red,... artistic, active, freedom, entail, apple fruit walking, jogging Person: {male, female}, {young, old}, {face, head-n-shoulder, wholebody}, {military, civilian}, From corpus and detection statistics Concept co-occurrence Detector complementarity beach, palm tree sky beach via color beach via texture 12

13 Multi-faceted Visual Taxonomy Augment tree-structure with facets Peer concepts within facets are mutually exclusive e.g. cat, dog; apple, banana; urban, rural Concepts across facets are not mutually exclusive e.g. baby, female, smiling Advantages Takes into account multiple-descriptions Can be used to infer positive and negative training data Still to be considered Co-occurring objects/scenes Common-sense relationships 13

14 Generative Models for Concept Structure Mutual exclusion Multiple facets Parentchild Cooccurrence Detector uncertainty 14

15 Generative Models for Concept Structure Mutual exclusion Multiple facets Parentchild Cooccurrence Detector uncertainty 15

16 Performance Comparisons 222-node taxonomy with six top-level facets (10~50 nodes, depth 3~7) Evaluate on 60,200 web images Baseline: ensemble SVM trained on another 240K images Mutual exclusion relationships are important: improves classification accuracy Multi-faceted tree structure (bnet) performs best in classification, and is robust to noise in concept prior. 16

17 Example Classification Results 17

18 An Image =? an image = one thousand words -- Which thousand words? -- Which hundred words do we use first? night, city, building, travel, urban, architecture, sculpture, lights, cloud, sky, reflection, artistic, ring, nikon, seoul, korea, perspective, ring,... We would like to use visual categories that are: Popular Visually observable and distinct Not available in camera metadata Not subjective or abstract urban korea nikon artistic 18

19 The Tag Value Problem A typical image tagging algorithm [Xie et. al. CIVR 10] Which k tags for each image The most accurate ones? The most specific ones? the most useful ones? Estimate factor of the following c -- visual tag u -- user-assigned social tag Outdoors? Sky? Ship? Red? Grass? Flower? Water? 19

20 Four tag re-ranking methods User-perceived value Information content Flickr popularity Tag posterior can we take into account both concept prior and tag popularity? 20

21 Tag posterior Goal: express tag confidence in terms of both classifier score, concept prior, and tag popularity.. Conditional independence Tag posterior factor: The use of all reranking factors: 21

22 Evaluation setup Recruited use group 24 users in US and UK Diverse professional backgrounds (R&D, marketing, services ) From casual photographers to amateur enthusiasts 5,998 user contributed photos uploaded to Flickr User evaluations Background questions Assigned subjective score to 60 visual tags. Label photos, delete incorrect machine-tags Subset of 5 users participated in pair-wise tag ranking study 22

23 User tolerance of tagging precision: 74~82% Mean: 2.59 Target Precision Mean: 74% Std. Dev.: 0.59 Target Precision Range: [68%, 80%] Mean: 4.57 Target Precision Mean: 82% Std. Dev.: 2.06 Target Precision Range: [72%, 87%] 23

24 Tag accuracy evaluation ImageNet-1000, UIUC-NEC Social 20 KNN-voting [Li, Snoek 09] ImageNet-1000, KNN ImageNet-1000, liblin* Aggregated performance over 50 core visual categories (excluding 11 colour categories). 24

25 User study: comparing tags Comparative user study to vote on two sets of tags. Flickr popularity and Posterior popularity are the winners. IC appears to be less useful due to the difficulty of reliably estimating tag prior. Weighting flickr popularity with IC still helps. Data-driven methods out-perform user perceived popularity 25

26 User Score vs Pair-wise Comparison Some rankings confirm intuition e.g. specific tags are preferred over general ones (beach > nature, mountain > outdoors) Some result from other factors of perceived usefulness e.g. common words are preferred (nature > vegetation) 26

27 Tags uploaded to Flickr.. 27

28 Two Approaches for Manual Concept Extraction [Yan, Natsev and Campbell, CVPR 08] Tagging -- widely used Associate a single image / video at a time with multiple keywords Social tagging by millions of users Browsing -- specific domains Associate multiple image / video with a single given keyword Suitable for domain experts IBM EVA [MM 05] CMU Extreme Retriever [MM 06] 28

29 Models for Annotation Time Tagging Time for tag f Start-up cost Browsing Time for judging positive Time for judging nega t f = 6.8 sec, t s = 5.6 sec t p = 1.4sec, t n = 0.2sec 29

30 Frequency-based Hybrid Annotation Combining both tagging and browsing interfaces to optimize the annotation time for manually annotating image collections Choose the tagging interface for infrequent keywords Choose the browsing interface for frequent keywords Formally model the annotation time as a function of word frequency, time per word, and annotation interfaces 30

31 Results for Frequency-based Annotation Results on two large collections: TRECVID and Corel More accurate than automatic annotation (100% accurate) More efficient than tagging / browsing annotation (2-x speedup) More effective than tagging / browsing in a given amount of time TRECVID Collection Corel Collection 31 31

32 Summary Find snow capped mountains Outdoors? Sky? Ship? Red? Grass? Flower? Water? 32

33 Thank you! We sincerely thank US Gov, IBM and client partners Ambreen Javed for data annotation Our fantastic group of users All of you for listening! Questions + more info: Lexing Xie <xlx@us.ibm.com> Demos + download

Columbia University High-Level Feature Detection: Parts-based Concept Detectors

Columbia University High-Level Feature Detection: Parts-based Concept Detectors TRECVID 2005 Workshop Columbia University High-Level Feature Detection: Parts-based Concept Detectors Dong-Qing Zhang, Shih-Fu Chang, Winston Hsu, Lexin Xie, Eric Zavesky Digital Video and Multimedia Lab

More information

Columbia University TRECVID-2005 Video Search and High-Level Feature Extraction

Columbia University TRECVID-2005 Video Search and High-Level Feature Extraction Columbia University TRECVID-2005 Video Search and High-Level Feature Extraction Shih-Fu Chang, Winston Hsu, Lyndon Kennedy, Lexing Xie, Akira Yanagawa, Eric Zavesky, Dong-Qing Zhang Digital Video and Multimedia

More information

A Reranking Approach for Context-based Concept Fusion in Video Indexing and Retrieval

A Reranking Approach for Context-based Concept Fusion in Video Indexing and Retrieval A Reranking Approach for Context-based Concept Fusion in Video Indexing and Retrieval Lyndon S. Kennedy Dept. of Electrical Engineering Columbia University New York, NY 10027 lyndon@ee.columbia.edu Shih-Fu

More information

The Stanford/Technicolor/Fraunhofer HHI Video Semantic Indexing System

The Stanford/Technicolor/Fraunhofer HHI Video Semantic Indexing System The Stanford/Technicolor/Fraunhofer HHI Video Semantic Indexing System Our first participation on the TRECVID workshop A. F. de Araujo 1, F. Silveira 2, H. Lakshman 3, J. Zepeda 2, A. Sheth 2, P. Perez

More information

Utilizing Semantic Word Similarity Measures for Video Retrieval

Utilizing Semantic Word Similarity Measures for Video Retrieval Utilizing Semantic Word Similarity Measures for Video Retrieval Yusuf Aytar Computer Vision Lab, University of Central Florida yaytar@cs.ucf.edu Mubarak Shah Computer Vision Lab, University of Central

More information

IBM Research TRECVID-2007 Video Retrieval System

IBM Research TRECVID-2007 Video Retrieval System IBM Research TRECVID-2007 Video Retrieval System Murray Campbell, Alexander Haubold, Ming Liu, Apostol (Paul) Natsev, John R. Smith, Jelena Tešić, Lexing Xie, Rong Yan, Jun Yang Abstract In this paper,

More information

ACM MM Dong Liu, Shuicheng Yan, Yong Rui and Hong-Jiang Zhang

ACM MM Dong Liu, Shuicheng Yan, Yong Rui and Hong-Jiang Zhang ACM MM 2010 Dong Liu, Shuicheng Yan, Yong Rui and Hong-Jiang Zhang Harbin Institute of Technology National University of Singapore Microsoft Corporation Proliferation of images and videos on the Internet

More information

User Strategies in Video Retrieval: a Case Study

User Strategies in Video Retrieval: a Case Study User Strategies in Video Retrieval: a Case Study L. Hollink 1, G.P. Nguyen 2, D.C. Koelma 2, A.Th. Schreiber 1, M. Worring 2 1 Business Informatics, Free University Amsterdam. {hollink,schreiber}@cs.vu.nl

More information

COMP90042 LECTURE 3 LEXICAL SEMANTICS COPYRIGHT 2018, THE UNIVERSITY OF MELBOURNE

COMP90042 LECTURE 3 LEXICAL SEMANTICS COPYRIGHT 2018, THE UNIVERSITY OF MELBOURNE COMP90042 LECTURE 3 LEXICAL SEMANTICS SENTIMENT ANALYSIS REVISITED 2 Bag of words, knn classifier. Training data: This is a good movie.! This is a great movie.! This is a terrible film. " This is a wonderful

More information

YUSUF AYTAR B.S. Ege University

YUSUF AYTAR B.S. Ege University SEMANTIC VIDEO RETRIEVAL USING HIGH LEVEL CONTEXT by YUSUF AYTAR B.S. Ege University A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in the School of Electrical

More information

Dissertation: Analysis, Indexing and Visualization of Presentation Videos

Dissertation: Analysis, Indexing and Visualization of Presentation Videos MICHELE MERLER IBM TJ Watson Research Center 19 Skyline Drive Hawthorne, NY, 10532 phone: (646) 510-1702 email: michele.merler@gmail.com www.cs.columbia.edu/~mmerler EDUCATION 2012 : Columbia University,

More information

Class 5: Attributes and Semantic Features

Class 5: Attributes and Semantic Features Class 5: Attributes and Semantic Features Rogerio Feris, Feb 21, 2013 EECS 6890 Topics in Information Processing Spring 2013, Columbia University http://rogerioferis.com/visualrecognitionandsearch Project

More information

Visual Query Suggestion

Visual Query Suggestion Visual Query Suggestion Zheng-Jun Zha, Linjun Yang, Tao Mei, Meng Wang, Zengfu Wang University of Science and Technology of China Textual Visual Query Suggestion Microsoft Research Asia Motivation Framework

More information

User Strategies in Video Retrieval: a Case Study

User Strategies in Video Retrieval: a Case Study User Strategies in Video Retrieval: a Case Study L. Hollink 1, G.P. Nguyen 2, D.C. Koelma 2, A.Th. Schreiber 1, M. Worring 2 1 Section Business Informatics Free University Amsterdam De Boelelaan 1081a

More information

Multimedia Information Retrieval

Multimedia Information Retrieval Multimedia Information Retrieval Prof Stefan Rüger Multimedia and Information Systems Knowledge Media Institute The Open University http://kmi.open.ac.uk/mmis Multimedia Information Retrieval 1. What are

More information

A Text-Based Approach to the ImageCLEF 2010 Photo Annotation Task

A Text-Based Approach to the ImageCLEF 2010 Photo Annotation Task A Text-Based Approach to the ImageCLEF 2010 Photo Annotation Task Wei Li, Jinming Min, Gareth J. F. Jones Center for Next Generation Localisation School of Computing, Dublin City University Dublin 9, Ireland

More information

Harvesting collective Images for Bi-Concept exploration

Harvesting collective Images for Bi-Concept exploration Harvesting collective Images for Bi-Concept exploration B.Nithya priya K.P.kaliyamurthie Abstract--- Noised positive as well as instructive pessimistic research examples commencing the communal web, to

More information

NATURAL LANGUAGE PROCESSING

NATURAL LANGUAGE PROCESSING NATURAL LANGUAGE PROCESSING LESSON 9 : SEMANTIC SIMILARITY OUTLINE Semantic Relations Semantic Similarity Levels Sense Level Word Level Text Level WordNet-based Similarity Methods Hybrid Methods Similarity

More information

CuZero: Embracing the Frontier of Interactive Visual Search for Informed Users

CuZero: Embracing the Frontier of Interactive Visual Search for Informed Users CuZero: Embracing the Frontier of Interactive Visual Search for Informed Users Eric Zavesky Shih-Fu Chang Dept. of Electrical Engineering Columbia University 1312 S.W.Mudd, 500 W. 120th St, New York, NY

More information

Leveraging flickr images for object detection

Leveraging flickr images for object detection Leveraging flickr images for object detection Elisavet Chatzilari Spiros Nikolopoulos Yiannis Kompatsiaris Outline Introduction to object detection Our proposal Experiments Current research 2 Introduction

More information

High-level Event Recognition in Internet Videos

High-level Event Recognition in Internet Videos High-level Event Recognition in Internet Videos Yu-Gang Jiang School of Computer Science Fudan University, Shanghai, China ygj@fudan.edu.cn Joint work with Guangnan Ye 1, Subh Bhattacharya 2, Dan Ellis

More information

Semantic Video Indexing

Semantic Video Indexing Semantic Video Indexing T-61.6030 Multimedia Retrieval Stevan Keraudy stevan.keraudy@tkk.fi Helsinki University of Technology March 14, 2008 What is it? Query by keyword or tag is common Semantic Video

More information

Video search requires efficient annotation of video content To some extent this can be done automatically

Video search requires efficient annotation of video content To some extent this can be done automatically VIDEO ANNOTATION Market Trends Broadband doubling over next 3-5 years Video enabled devices are emerging rapidly Emergence of mass internet audience Mainstream media moving to the Web What do we search

More information

Experimenting VIREO-374: Bag-of-Visual-Words and Visual-Based Ontology for Semantic Video Indexing and Search

Experimenting VIREO-374: Bag-of-Visual-Words and Visual-Based Ontology for Semantic Video Indexing and Search Experimenting VIREO-374: Bag-of-Visual-Words and Visual-Based Ontology for Semantic Video Indexing and Search Chong-Wah Ngo, Yu-Gang Jiang, Xiaoyong Wei Feng Wang, Wanlei Zhao, Hung-Khoon Tan and Xiao

More information

IJREAT International Journal of Research in Engineering & Advanced Technology, Volume 1, Issue 5, Oct-Nov, 2013 ISSN:

IJREAT International Journal of Research in Engineering & Advanced Technology, Volume 1, Issue 5, Oct-Nov, 2013 ISSN: Semi Automatic Annotation Exploitation Similarity of Pics in i Personal Photo Albums P. Subashree Kasi Thangam 1 and R. Rosy Angel 2 1 Assistant Professor, Department of Computer Science Engineering College,

More information

Rushes Video Segmentation Using Semantic Features

Rushes Video Segmentation Using Semantic Features Rushes Video Segmentation Using Semantic Features Athina Pappa, Vasileios Chasanis, and Antonis Ioannidis Department of Computer Science and Engineering, University of Ioannina, GR 45110, Ioannina, Greece

More information

Associating video frames with text

Associating video frames with text Associating video frames with text Pinar Duygulu and Howard Wactlar Informedia Project School of Computer Science University Informedia Digital Video Understanding Project IDVL interface returned for "El

More information

SEARCHING pictures on smart phones, PCs, and the

SEARCHING pictures on smart phones, PCs, and the IEEE TRANSACTIONS ON MULTIMEDIA, VOL. 14, NO. 4, AUGUST 2012 1091 Harvesting Social Images for Bi-Concept Search Xirong Li, Cees G. M. Snoek, Senior Member, IEEE, Marcel Worring, Member, IEEE, and Arnold

More information

Kristina Lerman University of Southern California. This lecture is partly based on slides prepared by Anon Plangprasopchok

Kristina Lerman University of Southern California. This lecture is partly based on slides prepared by Anon Plangprasopchok Kristina Lerman University of Southern California This lecture is partly based on slides prepared by Anon Plangprasopchok Social Web is a platform for people to create, organize and share information Users

More information

Consumer Video Understanding

Consumer Video Understanding Consumer Video Understanding A Benchmark Database + An Evaluation of Human & Machine Performance Yu-Gang Jiang, Guangnan Ye, Shih-Fu Chang, Daniel Ellis, Alexander C. Loui Columbia University Kodak Research

More information

Learning the Semantics of Multimedia Queries and Concepts from a Small Number of Examples

Learning the Semantics of Multimedia Queries and Concepts from a Small Number of Examples Learning the Semantics of Multimedia Queries and Concepts from a Small Number of Examples Apostol (Paul) Natsev IBM Watson Research Center 19 Skyline Drive Hawthorne, NY 10532 natsev@us.ibm.com Milind

More information

arxiv: v1 [cs.mm] 12 Jan 2016

arxiv: v1 [cs.mm] 12 Jan 2016 Learning Subclass Representations for Visually-varied Image Classification Xinchao Li, Peng Xu, Yue Shi, Martha Larson, Alan Hanjalic Multimedia Information Retrieval Lab, Delft University of Technology

More information

Easy Samples First: Self-paced Reranking for Zero-Example Multimedia Search

Easy Samples First: Self-paced Reranking for Zero-Example Multimedia Search Easy Samples First: Self-paced Reranking for Zero-Example Multimedia Search Lu Jiang 1, Deyu Meng 2, Teruko Mitamura 1, Alexander G. Hauptmann 1 1 School of Computer Science, Carnegie Mellon University

More information

AUTOMATIC VISUAL CONCEPT DETECTION IN VIDEOS

AUTOMATIC VISUAL CONCEPT DETECTION IN VIDEOS AUTOMATIC VISUAL CONCEPT DETECTION IN VIDEOS Nilam B. Lonkar 1, Dinesh B. Hanchate 2 Student of Computer Engineering, Pune University VPKBIET, Baramati, India Computer Engineering, Pune University VPKBIET,

More information

Multimedia Information Retrieval The case of video

Multimedia Information Retrieval The case of video Multimedia Information Retrieval The case of video Outline Overview Problems Solutions Trends and Directions Multimedia Information Retrieval Motivation With the explosive growth of digital media data,

More information

Integrating Semantic and Visual Facets for Browsing Digital Photo Collections

Integrating Semantic and Visual Facets for Browsing Digital Photo Collections Integrating Semantic and Visual Facets for Browsing Digital Photo Collections Ilaria Bartolini and Paolo Ciaccia DEIS, Università di Bologna - Italy {i.bartolini,paolo.ciaccia}@unibo.it Abstract. Managing

More information

Ontology-Based Web Query Classification for Research Paper Searching

Ontology-Based Web Query Classification for Research Paper Searching Ontology-Based Web Query Classification for Research Paper Searching MyoMyo ThanNaing University of Technology(Yatanarpon Cyber City) Mandalay,Myanmar Abstract- In web search engines, the retrieval of

More information

Boolean Queries. Keywords combined with Boolean operators:

Boolean Queries. Keywords combined with Boolean operators: Query Languages 1 Boolean Queries Keywords combined with Boolean operators: OR: (e 1 OR e 2 ) AND: (e 1 AND e 2 ) BUT: (e 1 BUT e 2 ) Satisfy e 1 but not e 2 Negation only allowed using BUT to allow efficient

More information

TERM BASED WEIGHT MEASURE FOR INFORMATION FILTERING IN SEARCH ENGINES

TERM BASED WEIGHT MEASURE FOR INFORMATION FILTERING IN SEARCH ENGINES TERM BASED WEIGHT MEASURE FOR INFORMATION FILTERING IN SEARCH ENGINES Mu. Annalakshmi Research Scholar, Department of Computer Science, Alagappa University, Karaikudi. annalakshmi_mu@yahoo.co.in Dr. A.

More information

Multimedia content continues to grow rapidly.

Multimedia content continues to grow rapidly. Standards John R. Smith IBM Large-Scale Concept Ontology for Multimedia Milind Naphade, John R. Smith, and Jelena Tesic IBM T.J. Watson Research Center Shih-Fu Chang, Winston Hsu, amd Lyndon Kennedy Columbia

More information

VIDEO has become the medium of choice in applications

VIDEO has become the medium of choice in applications IEEE TRANSACTIONS ON MULTIMEDIA, VOL. 9, NO. 5, AUGUST 2007 975 Adding Semantics to Detectors for Video Retrieval Cees G. M. Snoek, Member, IEEE, Bouke Huurnink, Laura Hollink, Maarten de Rijke, Guus Schreiber,

More information

Fast and Accurate Content-based Semantic Search in 100M Internet Videos

Fast and Accurate Content-based Semantic Search in 100M Internet Videos Fast and Accurate Content-based Semantic Search in 100M Internet Videos Lu Jiang 1, Shoou-I Yu 1, Deyu Meng 2, Yi Yang 3, Teruko Mitamura 1, Alexander Hauptmann 1 1 Carnegie Mellon University 2 Xi an Jiaotong

More information

Experiments of Image Retrieval Using Weak Attributes

Experiments of Image Retrieval Using Weak Attributes Columbia University Computer Science Department Technical Report # CUCS 005-12 (2012) Experiments of Image Retrieval Using Weak Attributes Felix X. Yu, Rongrong Ji, Ming-Hen Tsai, Guangnan Ye, Shih-Fu

More information

Tansu Alpcan C. Bauckhage S. Agarwal

Tansu Alpcan C. Bauckhage S. Agarwal 1 / 16 C. Bauckhage S. Agarwal Deutsche Telekom Laboratories GBR 2007 2 / 16 Outline 3 / 16 Overview A novel expert peering system for community-based information exchange A graph-based scheme consisting

More information

Information Retrieval and Web Search

Information Retrieval and Web Search Information Retrieval and Web Search Relevance Feedback. Query Expansion Instructor: Rada Mihalcea Intelligent Information Retrieval 1. Relevance feedback - Direct feedback - Pseudo feedback 2. Query expansion

More information

Semantic Web. Ontology Engineering and Evaluation. Morteza Amini. Sharif University of Technology Fall 93-94

Semantic Web. Ontology Engineering and Evaluation. Morteza Amini. Sharif University of Technology Fall 93-94 ه عا ی Semantic Web Ontology Engineering and Evaluation Morteza Amini Sharif University of Technology Fall 93-94 Outline Ontology Engineering Class and Class Hierarchy Ontology Evaluation 2 Outline Ontology

More information

Multimedia Data Management M

Multimedia Data Management M ALMA MATER STUDIORUM - UNIVERSITÀ DI BOLOGNA Multimedia Data Management M Second cycle degree programme (LM) in Computer Engineering University of Bologna Semantic Multimedia Data Annotation Home page:

More information

Searching Visual Semantic Spaces with Concept Filters

Searching Visual Semantic Spaces with Concept Filters Searching Visual Semantic Spaces with Concept Filters Eric Zavesky, Zhu Liu, David Gibbon, Behzad Shahraray AT&T Labs Research, Middletown, NJ {ezavesky, zliu, dcg, behzad}@research.att.com Abstract Semantic

More information

Part-based and local feature models for generic object recognition

Part-based and local feature models for generic object recognition Part-based and local feature models for generic object recognition May 28 th, 2015 Yong Jae Lee UC Davis Announcements PS2 grades up on SmartSite PS2 stats: Mean: 80.15 Standard Dev: 22.77 Vote on piazza

More information

Exploiting noisy web data for largescale visual recognition

Exploiting noisy web data for largescale visual recognition Exploiting noisy web data for largescale visual recognition Lamberto Ballan University of Padova, Italy CVPRW WebVision - Jul 26, 2017 Datasets drive computer vision progress ImageNet Slide credit: O.

More information

The Fraunhofer IDMT at ImageCLEF 2011 Photo Annotation Task

The Fraunhofer IDMT at ImageCLEF 2011 Photo Annotation Task The Fraunhofer IDMT at ImageCLEF 2011 Photo Annotation Task Karolin Nagel, Stefanie Nowak, Uwe Kühhirt and Kay Wolter Fraunhofer Institute for Digital Media Technology (IDMT) Ehrenbergstr. 31, 98693 Ilmenau,

More information

WEIGHTING QUERY TERMS USING WORDNET ONTOLOGY

WEIGHTING QUERY TERMS USING WORDNET ONTOLOGY IJCSNS International Journal of Computer Science and Network Security, VOL.9 No.4, April 2009 349 WEIGHTING QUERY TERMS USING WORDNET ONTOLOGY Mohammed M. Sakre Mohammed M. Kouta Ali M. N. Allam Al Shorouk

More information

Semantic Extensions to Syntactic Analysis of Queries Ben Handy, Rohini Rajaraman

Semantic Extensions to Syntactic Analysis of Queries Ben Handy, Rohini Rajaraman Semantic Extensions to Syntactic Analysis of Queries Ben Handy, Rohini Rajaraman Abstract We intend to show that leveraging semantic features can improve precision and recall of query results in information

More information

MULTIVIEW RANK LEARNING FOR MULTIMEDIA KNOWN ITEM SEARCH

MULTIVIEW RANK LEARNING FOR MULTIMEDIA KNOWN ITEM SEARCH MULTIVIEW RANK LEARNING FOR MULTIMEDIA KNOWN ITEM SEARCH by David L Etter A Dissertation Submitted to the Graduate Faculty of George Mason University In Partial fulfillment of The Requirements for the

More information

Intelligent ranking for photo galleries using sharing intent

Intelligent ranking for photo galleries using sharing intent Technical Disclosure Commons Defensive Publications Series June 20, 2018 Intelligent ranking for photo galleries using sharing intent John Oberbeck Follow this and additional works at: https://www.tdcommons.org/dpubs_series

More information

AS DIGITAL cameras become more affordable and widespread,

AS DIGITAL cameras become more affordable and widespread, IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 17, NO. 3, MARCH 2008 407 Integrating Concept Ontology and Multitask Learning to Achieve More Effective Classifier Training for Multilevel Image Annotation Jianping

More information

Semantic Web Systems Ontologies Jacques Fleuriot School of Informatics

Semantic Web Systems Ontologies Jacques Fleuriot School of Informatics Semantic Web Systems Ontologies Jacques Fleuriot School of Informatics 15 th January 2015 In the previous lecture l What is the Semantic Web? Web of machine-readable data l Aims of the Semantic Web Automated

More information

Building a Facebook application to investigate how people tag pictures

Building a Facebook application to investigate how people tag pictures Building a Facebook application to investigate how people tag pictures Victor Hartanto Wibisono (U4644427) supervised by: Dr. Lexing Xie The Problem Victor s Friend Owner John Stranger Victor Sam COMP3750:

More information

NUS-WIDE: A Real-World Web Image Database from National University of Singapore

NUS-WIDE: A Real-World Web Image Database from National University of Singapore NUS-WIDE: A Real-World Web Image Database from National University of Singapore Tat-Seng Chua, Jinhui Tang, Richang Hong, Haojie Li, Zhiping Luo, Yantao Zheng National University of Singapore Computing

More information

Semantic Context Transfer across Heterogeneous Sources for Domain Adaptive Video Search

Semantic Context Transfer across Heterogeneous Sources for Domain Adaptive Video Search Semantic Context Transfer across Heterogeneous Sources for Domain Adaptive Video Search Yu-Gang Jiang, Chong-Wah Ngo, Shih-Fu Chang Department of Computer Science, City University of Hong Kong Department

More information

Video annotation based on adaptive annular spatial partition scheme

Video annotation based on adaptive annular spatial partition scheme Video annotation based on adaptive annular spatial partition scheme Guiguang Ding a), Lu Zhang, and Xiaoxu Li Key Laboratory for Information System Security, Ministry of Education, Tsinghua National Laboratory

More information

Automatic Image Annotation and Retrieval Using Hybrid Approach

Automatic Image Annotation and Retrieval Using Hybrid Approach Automatic Image Annotation and Retrieval Using Hybrid Approach Zhixin Li, Weizhong Zhao 2, Zhiqing Li 2, Zhiping Shi 3 College of Computer Science and Information Technology, Guangxi Normal University,

More information

Xian-Sheng Hua ( 华先胜 )

Xian-Sheng Hua ( 华先胜 ) Machine Learning and Applications Workshop 2009 Xian-Sheng Hua ( 华先胜 ) Lead Researcher, Media Computing Group, Microsoft Research Asia Nanjing China Nov 7-8, 2009 Introduce research problems in multimedia

More information

USING ASSOCIATION RULE MINING TO ENRICH SEMANTIC CONCEPTS FOR VIDEO RETRIEVAL

USING ASSOCIATION RULE MINING TO ENRICH SEMANTIC CONCEPTS FOR VIDEO RETRIEVAL USING ASSOCIATION RULE MINING TO ENRICH SEMANTIC CONCEPTS FOR VIDEO RETRIEVAL Nastaran Fatemi, Florian Poulin, Laura E. Raileanu HEIG-VD, Univ. of Applied Science of Western Switzerland,Yverdon-les-Bains,

More information

Interactive Image Search with Attributes

Interactive Image Search with Attributes Interactive Image Search with Attributes Adriana Kovashka Department of Computer Science January 13, 2015 Joint work with Kristen Grauman and Devi Parikh We Need Search to Access Visual Data 144,000 hours

More information

BUAA AUDR at ImageCLEF 2012 Photo Annotation Task

BUAA AUDR at ImageCLEF 2012 Photo Annotation Task BUAA AUDR at ImageCLEF 2012 Photo Annotation Task Lei Huang, Yang Liu State Key Laboratory of Software Development Enviroment, Beihang University, 100191 Beijing, China huanglei@nlsde.buaa.edu.cn liuyang@nlsde.buaa.edu.cn

More information

A Survey Of Different Text Mining Techniques Varsha C. Pande 1 and Dr. A.S. Khandelwal 2

A Survey Of Different Text Mining Techniques Varsha C. Pande 1 and Dr. A.S. Khandelwal 2 A Survey Of Different Text Mining Techniques Varsha C. Pande 1 and Dr. A.S. Khandelwal 2 1 Department of Electronics & Comp. Sc, RTMNU, Nagpur, India 2 Department of Computer Science, Hislop College, Nagpur,

More information

Automatic Discovery of Query-Class-Dependent Models for Multimodal Search

Automatic Discovery of Query-Class-Dependent Models for Multimodal Search Automatic Discovery of -Class-Dependent Models for Multimodal Search Lyndon S. Kennedy Dept. of Electrical Engineering Columbia University New York, NY 10027 lyndon@ee.columbia.edu Apostol (Paul) Natsev

More information

Question Answering Systems

Question Answering Systems Question Answering Systems An Introduction Potsdam, Germany, 14 July 2011 Saeedeh Momtazi Information Systems Group Outline 2 1 Introduction Outline 2 1 Introduction 2 History Outline 2 1 Introduction

More information

CS 1674: Intro to Computer Vision. Attributes. Prof. Adriana Kovashka University of Pittsburgh November 2, 2016

CS 1674: Intro to Computer Vision. Attributes. Prof. Adriana Kovashka University of Pittsburgh November 2, 2016 CS 1674: Intro to Computer Vision Attributes Prof. Adriana Kovashka University of Pittsburgh November 2, 2016 Plan for today What are attributes and why are they useful? (paper 1) Attributes for zero-shot

More information

Tag Based Image Search by Social Re-ranking

Tag Based Image Search by Social Re-ranking Tag Based Image Search by Social Re-ranking Vilas Dilip Mane, Prof.Nilesh P. Sable Student, Department of Computer Engineering, Imperial College of Engineering & Research, Wagholi, Pune, Savitribai Phule

More information

Quaero at TRECVid 2012: Semantic Indexing

Quaero at TRECVid 2012: Semantic Indexing Quaero at TRECVid 2012: Semantic Indexing Bahjat Safadi 1, Nadia Derbas 1, Abdelkader Hamadi 1, Franck Thollard 1, Georges Quénot 1, Jonathan Delhumeau 2, Hervé Jégou 2, Tobias Gehrig 3, Hazim Kemal Ekenel

More information

CHAPTER 5 SEARCH ENGINE USING SEMANTIC CONCEPTS

CHAPTER 5 SEARCH ENGINE USING SEMANTIC CONCEPTS 82 CHAPTER 5 SEARCH ENGINE USING SEMANTIC CONCEPTS In recent years, everybody is in thirst of getting information from the internet. Search engines are used to fulfill the need of them. Even though the

More information

IEEE TRANSACTIONS ON MULTIMEDIA, VOL. 14, NO. 2, APRIL Efficient Genre-Specific Semantic Video Indexing

IEEE TRANSACTIONS ON MULTIMEDIA, VOL. 14, NO. 2, APRIL Efficient Genre-Specific Semantic Video Indexing IEEE TRANSACTIONS ON MULTIMEDIA, VOL. 14, NO. 2, APRIL 2012 291 Efficient Genre-Specific Semantic Video Indexing Jun Wu and Marcel Worring, Member, IEEE Abstract Large video collections such as YouTube

More information

WordNet-based User Profiles for Semantic Personalization

WordNet-based User Profiles for Semantic Personalization PIA 2005 Workshop on New Technologies for Personalized Information Access WordNet-based User Profiles for Semantic Personalization Giovanni Semeraro, Marco Degemmis, Pasquale Lops, Ignazio Palmisano LACAM

More information

CHAPTER 5 EXPERT LOCATOR USING CONCEPT LINKING

CHAPTER 5 EXPERT LOCATOR USING CONCEPT LINKING 94 CHAPTER 5 EXPERT LOCATOR USING CONCEPT LINKING 5.1 INTRODUCTION Expert locator addresses the task of identifying the right person with the appropriate skills and knowledge. In large organizations, it

More information

Web Information Retrieval using WordNet

Web Information Retrieval using WordNet Web Information Retrieval using WordNet Jyotsna Gharat Asst. Professor, Xavier Institute of Engineering, Mumbai, India Jayant Gadge Asst. Professor, Thadomal Shahani Engineering College Mumbai, India ABSTRACT

More information

VERGE: A Video Interactive Retrieval Engine

VERGE: A Video Interactive Retrieval Engine VERGE: A Video Interactive Retrieval Engine Stefanos Vrochidis, Anastasia Moumtzidou, Paul King, Anastasios Dimou, Vasileios Mezaris and Ioannis Kompatsiaris Informatics and Telematics Institute 6th Km

More information

Refining Image Annotation Using Contextual Relations Between Words

Refining Image Annotation Using Contextual Relations Between Words Refining Image Annotation Using Contextual Relations Between Words ABSTRACT Yong Wang Department of Computer Science Queen Mary, University of London Mile End Road, London, UK, E1 4NS ywang@dcs.qmul.ac.uk

More information

Visual Search: 3 Levels of Real-Time Feedback

Visual Search: 3 Levels of Real-Time Feedback Visual Search: 3 Levels of Real-Time Feedback Prof. Shih-Fu Chang Department of Electrical Engineering Digital Video and Multimedia Lab http://www.ee.columbia.edu/dvmm A lot of work on Image Classification

More information

MICC-UNIFI at ImageCLEF 2013 Scalable Concept Image Annotation

MICC-UNIFI at ImageCLEF 2013 Scalable Concept Image Annotation MICC-UNIFI at ImageCLEF 2013 Scalable Concept Image Annotation Tiberio Uricchio, Marco Bertini, Lamberto Ballan, and Alberto Del Bimbo Media Integration and Communication Center (MICC) Università degli

More information

Information Technology for Documentary Data Representation

Information Technology for Documentary Data Representation ALMA MATER STUDIORUM - UNIVERSITÀ DI BOLOGNA Information Technology for Documentary Data Representation Laurea Magistrale in Scienze del Libro e del Documento University of Bologna Multimedia Information

More information

Ontology Matching with CIDER: Evaluation Report for the OAEI 2008

Ontology Matching with CIDER: Evaluation Report for the OAEI 2008 Ontology Matching with CIDER: Evaluation Report for the OAEI 2008 Jorge Gracia, Eduardo Mena IIS Department, University of Zaragoza, Spain {jogracia,emena}@unizar.es Abstract. Ontology matching, the task

More information

Visual Dictionary: Towards a Higher-level Visual Representation for Object Categorization. CHUA, Tat-Seng School of Computing

Visual Dictionary: Towards a Higher-level Visual Representation for Object Categorization. CHUA, Tat-Seng School of Computing Visual Dictionary: Towards a Higher-level Visual Representation for Object Categorization CHUA, Tat-Seng School of Computing Outline of Talk Introduction Current Approaches Towards Higher Level Visual

More information

Optimal Query. Assume that the relevant set of documents C r. 1 N C r d j. d j. Where N is the total number of documents.

Optimal Query. Assume that the relevant set of documents C r. 1 N C r d j. d j. Where N is the total number of documents. Optimal Query Assume that the relevant set of documents C r are known. Then the best query is: q opt 1 C r d j C r d j 1 N C r d j C r d j Where N is the total number of documents. Note that even this

More information

Data Modelling and. Multimedia. Databases M. Multimedia. Information Retrieval Part III. Outline

Data Modelling and. Multimedia. Databases M. Multimedia. Information Retrieval Part III. Outline ALMA MATER STUDIORUM - UNIVERSITÀ DI BOLOGNA Data Modelling and Multimedia Databases M International Second cycle degree programme (LM) in Digital Humanities and Digital Knowledge (DHDK) University of

More information

MEASURING SEMANTIC SIMILARITY BETWEEN WORDS AND IMPROVING WORD SIMILARITY BY AUGUMENTING PMI

MEASURING SEMANTIC SIMILARITY BETWEEN WORDS AND IMPROVING WORD SIMILARITY BY AUGUMENTING PMI MEASURING SEMANTIC SIMILARITY BETWEEN WORDS AND IMPROVING WORD SIMILARITY BY AUGUMENTING PMI 1 KAMATCHI.M, 2 SUNDARAM.N 1 M.E, CSE, MahaBarathi Engineering College Chinnasalem-606201, 2 Assistant Professor,

More information

Question Answering Approach Using a WordNet-based Answer Type Taxonomy

Question Answering Approach Using a WordNet-based Answer Type Taxonomy Question Answering Approach Using a WordNet-based Answer Type Taxonomy Seung-Hoon Na, In-Su Kang, Sang-Yool Lee, Jong-Hyeok Lee Department of Computer Science and Engineering, Electrical and Computer Engineering

More information

Web-Scale Image Search and Their Applications

Web-Scale Image Search and Their Applications Web-Scale Image Search and Their Applications Sung-Eui Yoon KAIST http://sglab.kaist.ac.kr Project Guidelines: Project Topics Any topics related to the course theme are okay You can find topics by browsing

More information

Search Engines. Information Retrieval in Practice

Search Engines. Information Retrieval in Practice Search Engines Information Retrieval in Practice All slides Addison Wesley, 2008 Beyond Bag of Words Bag of Words a document is considered to be an unordered collection of words with no relationships Extending

More information

Ontology Based Search Engine

Ontology Based Search Engine Ontology Based Search Engine K.Suriya Prakash / P.Saravana kumar Lecturer / HOD / Assistant Professor Hindustan Institute of Engineering Technology Polytechnic College, Padappai, Chennai, TamilNadu, India

More information

Baseball Game Highlight & Event Detection

Baseball Game Highlight & Event Detection Baseball Game Highlight & Event Detection Student: Harry Chao Course Adviser: Winston Hu 1 Outline 1. Goal 2. Previous methods 3. My flowchart 4. My methods 5. Experimental result 6. Conclusion & Future

More information

Open Research Online The Open University s repository of research publications and other research outputs

Open Research Online The Open University s repository of research publications and other research outputs Open Research Online The Open University s repository of research publications and other research outputs Using manual and automated annotations to search images by semantic similarity Journal Item How

More information

The Effectiveness of Concept Based Search for Video Retrieval

The Effectiveness of Concept Based Search for Video Retrieval The Effectiveness of Concept Based Search for Video Retrieval Claudia Hauff and Robin Aly and Djoerd Hiemstra Computer Science University Twente P.O. Box 217 7500 AE Enschede The Netherlands e-mail: {c.hauff,

More information

Text Mining. Munawar, PhD. Text Mining - Munawar, PhD

Text Mining. Munawar, PhD. Text Mining - Munawar, PhD 10 Text Mining Munawar, PhD Definition Text mining also is known as Text Data Mining (TDM) and Knowledge Discovery in Textual Database (KDT).[1] A process of identifying novel information from a collection

More information

Definition, Detection, and Evaluation of Meeting Events in Airport Surveillance Videos

Definition, Detection, and Evaluation of Meeting Events in Airport Surveillance Videos Definition, Detection, and Evaluation of Meeting Events in Airport Surveillance Videos Sung Chun Lee, Chang Huang, and Ram Nevatia University of Southern California, Los Angeles, CA 90089, USA sungchun@usc.edu,

More information

Lab for Media Search, National University of Singapore 1

Lab for Media Search, National University of Singapore 1 1 2 Word2Image: Towards Visual Interpretation of Words Haojie Li Introduction Motivation A picture is worth 1000 words Traditional dictionary Containing word entries accompanied by photos or drawing to

More information

Ontology Based Prediction of Difficult Keyword Queries

Ontology Based Prediction of Difficult Keyword Queries Ontology Based Prediction of Difficult Keyword Queries Lubna.C*, Kasim K Pursuing M.Tech (CSE)*, Associate Professor (CSE) MEA Engineering College, Perinthalmanna Kerala, India lubna9990@gmail.com, kasim_mlp@gmail.com

More information

Fuzzy Knowledge-based Image Annotation Refinement

Fuzzy Knowledge-based Image Annotation Refinement 284 Int'l Conf. IP, Comp. Vision, and Pattern Recognition IPCV'15 Fuzzy Knowledge-based Image Annotation Refinement M. Ivašić-Kos 1, M. Pobar 1 and S. Ribarić 2 1 Department of Informatics, University

More information

Making Sense Out of the Web

Making Sense Out of the Web Making Sense Out of the Web Rada Mihalcea University of North Texas Department of Computer Science rada@cs.unt.edu Abstract. In the past few years, we have witnessed a tremendous growth of the World Wide

More information