請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/37960
完整後設資料紀錄
DC 欄位 | 值 | 語言 |
---|---|---|
dc.contributor.advisor | 陳信希 | |
dc.contributor.author | Yih-Cheng Chang | en |
dc.contributor.author | 張亦塵 | zh_TW |
dc.date.accessioned | 2021-06-13T15:53:31Z | - |
dc.date.available | 2008-07-03 | |
dc.date.copyright | 2008-07-03 | |
dc.date.issued | 2008 | |
dc.date.submitted | 2008-06-19 | |
dc.identifier.citation | Banerjee, Satanjeev, T. Pedersen. (2003). “Extended gloss overlaps as a measure of semantic relatedness,” Proceedings of the 18th IJCAI. Acapulco, pp. 805-810.
D. Blei, and M. I. Jordan. (2003). “Modeling Annotated Data,” SIGIR, 2003. S. Brody, R. Navigli, M. Lapata. (2006). “Ensemble Methods for Unsupervised WSD,” Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the ACL, pp. 97–104. D. Cai, X. He, Z. Li, W.-Y. Ma, J.-R. Wen. (2004). “Hierarchical Clustering of WWW Image Search Results Using Visual, Textual and Link Information,” Proceeding of 12th ACM International Conference on Multimedia, New York City, USA, Oct. 2004. G. Carneiro, and N. Vasconcelos. (2005). “A Database Centric View of Semantic Image Annotation and Retrieval,” SIGIR, 2005. P. Duygulu, K. Barnard, N. Freitas, and D. Forsyth. (2002). “Object Recognition as Machine Translation: Learning a Lexicon for a Fixed Image Vocabulary”, ECCV, 2002, pp. 97-112. Galley, Michel, K. McKeown. (2003). “Improving word sense disambiguation in lexical chaining,” Proceedings of the 18th IJCAI. Acapulco, pp. 1486-1488. A. Ghoshal, P. Ircing, and S. Khudanpur. (2005). “HiddenMarkov Models for Automatic Annotation and Content-Based Retrieval of Images and Video,” SIGIR, 2005. J. Jeon, and R. Manmatha. (2004). “Automatic Image Annotation of News Images with Large Vocabularies and Low Quality Training Data,” ACM Multimedia, 2004. V. Lavrenko, R. Manmatha, and J. Jeon. (2003). “A Model for Learning the Semantics of Pictures,” NIPS, 2003. Lesk, Michael. (1986). “Automatic sense disambiguation using machine readable dictionaries: How to tell a pine cone from an ice cream cone,” Proceedings of the 5th SIGDOC. New York, NY, pp. 24-26. B.-T. Li, K. Goh, E. Chang. (2003). “Confidence-based Dynamic Ensemble for Image Annotation and Semantics Discovery,” ACM Multimedia, 2003, pp. 195-206. N. Loeff, C.O. Alm, D.A. Forsyth. (2006). “Discrimination image senses by clustering with multimodal features,” Proceedings of the COLING/ACL 2006 Main Conference Poster Sessions, pp. 547–554. McCarthy, Diana, R. Koeling, J. Weeds, J. Carroll. (2004). “Finding predominant senses in untagged text,” Proceedings of the 42th ACL. Barcelona, Spain, pp. 280-287. Navigli, Roberto, P. Velardi. (2005). “Structural semantic interconnections: a knowledge-based approach to word sense disambiguation,” PAMI 27(7), pp. 1075-1088. X.-J. Wang, L. Zhang, F. Jing, W.-Y. Ma. (2006). “AnnoSearch: Image Auto-Annotation by Search,” Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 2 (CVPR06), pp. 1482-1490. K. Yanai, K. Barnard. (2005). “Probabilistic web image gathering,” SIGMM, pp.57–64. | |
dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/37960 | - |
dc.description.abstract | 最近幾年,網路圖片的數量有爆炸性的成長,如何檢索網路上的圖片也變得越來越重要。圖義辨識這個研究是用來解決檢索網路圖片時會遇到的歧異性問題,這個題目可以用來提高檢索網路圖片的效率,或運用在圖像標註及物件辨識之訓練資料的搜集上。圖義辨識是一個新的,還沒被徹底研究過的議題,但未來可能會變得越來越重要。
在這篇論文中,我們將會分析和探討圖義辨識這個議題,因為網路上的圖義很多都沒有被包含在字典裡,所以我們提出了一個方法來找出網路上的圖義,對於找到的每個圖義,我們提出的方法能夠在不需要使用任何人力標記的情況下收集到和每個圖義相關的圖片及網頁,和之前的研究不同,我們採用分類而非分群的概念來處理圖義辨識這個議題,我們提出了四種圖義的分類器以及一個合併這些分類器的方法,並對實驗中的數個步驟進行評估和討論,在論文的最後我們會對這篇論文做個總結並探討一些未來能夠繼續研究的有趣議題。 | zh_TW |
dc.description.abstract | In these few years, images in the web have explosively increased. Image retrieval for web images becomes more and more important. Image sense disambiguation/discrimination (ISD) is a task to disambiguate/discriminate image senses of retrieved web images. This technology can be used to improve the performance of web image retrieval or be applied in image annotation or object recognition tasks to help collecting training samples. ISD is a new task not being well studied but may become important in the future.
In this thesis, we analyze and discuss ISD problem. We propose a method to find senses of web images. There may be many senses in the web are not be included in the dictionary. For each sense, we collect sample images and pages without human annotation. Unlike previous approaches that use clustering methods in ISD, we use classifying method instead. Four kinds of classifiers and a merge method are proposed in this thesis. The steps of our methods are evaluated and discussed and in the end of this thesis we will summarize our work and discuss some interesting future works. | en |
dc.description.provenance | Made available in DSpace on 2021-06-13T15:53:31Z (GMT). No. of bitstreams: 1 ntu-97-R93922041-1.pdf: 721012 bytes, checksum: 989ca585fd94807ca384f0e52d2ed4a5 (MD5) Previous issue date: 2008 | en |
dc.description.tableofcontents | CHAPTER 1 INTRODUCTION 1
1.1MOTIVATION 1 1.1.1 Ambiguity in Web Image Retrieval 1 1.1.2 Image Annotation and Object Recognition 2 1.2 RESEARCH PROBLEMS 3 1.2.1 Finding Image Senses 3 1.2.1.1 Is WordNet Enough? 3 1.2.1.2 Finding Samples for Each Sense without Human Annotation 4 1.2.2 Classifying Image Senses 5 1.3 ORGANIZATION OF THIS THESIS 6 CHAPTER 2 RELATED WORKS 7 2.1 IMAGE SENSE DISCRIMINATION 7 2.2 UNSUPERVISED WORD DISAMBIGUATION 8 2.3 IMAGE ANNOTATION 9 CHAPTER 3 IMAGE SENSE FINDING 11 3.1 CANDIDATE VOCABULARIES TO REPRESENT SENSES 11 3.1.1 Finding Candidate Vocabularies 11 3.1.2 Finding Stop Words 14 3.1.3 Finding Relationships 15 3.2 GROUPING 19 3.3 FINDING SAMPLES FOR EACH SENSE 22 3.3.1 Collecting Samples 22 3.3.2 Sense Filtering 25 3.4 EXPERIMENT AND DISCUSSION FOR IMAGE SENSE FINDING 27 3.4.1 Evaluation for Coverage 27 3.4.2 Evaluating the Precision of Image Samples 28 CHAPTER 4 CLASSIFYING IMAGE SENSES 34 4.1 OVERVIEW 34 4.1.1 Features for Samples 34 4.1.2 Features for Target Image 35 4.2 IMAGE SENSES CLASSIFIERS 35 4.2.1 Text Classifier 35 4.2.2 Image Classifier 36 4.2.3 Website Classifier 37 4.2.3 Expand Text Feature and Classifier 38 4.3 EXPERIMENTS 39 4.3.1 Preparation of Testing Data Set and Answer Keys 39 4.3.2 Performance Measures and Baseline 40 4.3.3 Evaluation for Text Classifier 42 4.3.4 Evaluation for Image Classifier 49 4.3.5 Evaluation for Website Classifier 53 4.3.6 Experiment results for text classifier 59 4.4 MERGE THE CLASSIFIERS 63 4.5 ANALYSIS 67 4.5.1 Performance of Different Queries 67 4.5.2 Comparison with Human 68 4.5.3 Measuring the Execution Time 69 CHAPTER 5 CONCLUSION AND FUTURE WORKS 71 5.1 SUMMARY OF ACHIEVEMENTS 71 5.2 FUTURE WORK 72 REFERENCES 73 | |
dc.language.iso | en | |
dc.title | 網路檢索圖片集之圖義辨識方法研究 | zh_TW |
dc.title | Image Sense Disambiguation in Web Image Retrieval | en |
dc.type | Thesis | |
dc.date.schoolyear | 96-2 | |
dc.description.degree | 碩士 | |
dc.contributor.oralexamcommittee | 鄭卜壬,張俊盛,梁婷 | |
dc.subject.keyword | 圖義辨識,圖義區分,網路圖片檢索,圖像標註,物件辨識, | zh_TW |
dc.subject.keyword | image sense disambiguation,image sense discrimination,web image retrieval,image annotation,object recognition, | en |
dc.relation.page | 75 | |
dc.rights.note | 有償授權 | |
dc.date.accepted | 2008-06-23 | |
dc.contributor.author-college | 電機資訊學院 | zh_TW |
dc.contributor.author-dept | 資訊工程學研究所 | zh_TW |
顯示於系所單位: | 資訊工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-97-1.pdf 目前未授權公開取用 | 704.11 kB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。