Pub Date : 2010-12-13DOI: 10.1109/VSMM.2010.5665964
Yuto Maruyama, Gamhewage C. de Silva, T. Yamasaki, K. Aizawa
This paper presents a method to classify food images by updating the model of Bayesian network incrementally. We have been investigating a “food log” system which makes use of image analysis, and it can automatically detect food images and estimate the food balance (using a simple nutrition model). It also enables users to easily modify the results of the analysis when they contain errors. So far, the system does not make use of the corrections made by the users to improve the performance of classification. In this paper, we propose to incrementally update the classifier based on Baysian network so that the results of analysis will be improved by using the user's corrections. With the incremental updating, the accuracy of food image detection is improved from 89% to 92%.
{"title":"Personalization of food image analysis","authors":"Yuto Maruyama, Gamhewage C. de Silva, T. Yamasaki, K. Aizawa","doi":"10.1109/VSMM.2010.5665964","DOIUrl":"https://doi.org/10.1109/VSMM.2010.5665964","url":null,"abstract":"This paper presents a method to classify food images by updating the model of Bayesian network incrementally. We have been investigating a “food log” system which makes use of image analysis, and it can automatically detect food images and estimate the food balance (using a simple nutrition model). It also enables users to easily modify the results of the analysis when they contain errors. So far, the system does not make use of the corrections made by the users to improve the performance of classification. In this paper, we propose to incrementally update the classifier based on Baysian network so that the results of analysis will be improved by using the user's corrections. With the incremental updating, the accuracy of food image detection is improved from 89% to 92%.","PeriodicalId":348792,"journal":{"name":"2010 16th International Conference on Virtual Systems and Multimedia","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127019338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-12-13DOI: 10.1109/VSMM.2010.5665970
F. Chen, Wei Wang
Activity recognition is one of the most challenging problems in the video-based surveillance and computer-vision. In this paper we propose a novel approach to recognize human activity in which we decompose an activity into multiple stochastic processes, each corresponding to one scale of motion details. We present a hierarchical durational-state dynamic Bayesian network(HDS-DBN) to model two stochastic processes which are related to two appropriate scales in intelligent surveillance. In this approach the features we extracted are divided into two classes: global features and local features, which are at two different spatial scales. The HDS-DBN model structure combines global features with local ones harmoniously. The effectiveness of our approach is demonstrated by the experiments.
{"title":"Activity recognition through multi-scale dynamic Bayesian network","authors":"F. Chen, Wei Wang","doi":"10.1109/VSMM.2010.5665970","DOIUrl":"https://doi.org/10.1109/VSMM.2010.5665970","url":null,"abstract":"Activity recognition is one of the most challenging problems in the video-based surveillance and computer-vision. In this paper we propose a novel approach to recognize human activity in which we decompose an activity into multiple stochastic processes, each corresponding to one scale of motion details. We present a hierarchical durational-state dynamic Bayesian network(HDS-DBN) to model two stochastic processes which are related to two appropriate scales in intelligent surveillance. In this approach the features we extracted are divided into two classes: global features and local features, which are at two different spatial scales. The HDS-DBN model structure combines global features with local ones harmoniously. The effectiveness of our approach is demonstrated by the experiments.","PeriodicalId":348792,"journal":{"name":"2010 16th International Conference on Virtual Systems and Multimedia","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128178931","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-12-13DOI: 10.1109/VSMM.2010.5665963
K. Aizawa, Gamhewage C. de Silva, Makoto Ogawa, Yohei Sato
We present the current status of FoodLog, a multimedia Internet application that enables easy capture and archival of information regarding our daily meals. The primary purpose of FoodLog is to facilitate dietary management support with minimum manual recording of information. It analyzes image archives that belong to a user to identify images of meals. Further image analysis determines the nutritional composition of these meals and stores the data to form a log. The user can view the data from this log in different formats, and also edit the data to correct any mistakes that occurred during image analysis. This application was recently opened to the public, and had accumulated approximately 25000 images during the first two months since its launch. We present the current status of this application, and discuss our future plans to extend it to allow interaction between users and more effective dietary management.
{"title":"Food Log by snapping and processing images","authors":"K. Aizawa, Gamhewage C. de Silva, Makoto Ogawa, Yohei Sato","doi":"10.1109/VSMM.2010.5665963","DOIUrl":"https://doi.org/10.1109/VSMM.2010.5665963","url":null,"abstract":"We present the current status of FoodLog, a multimedia Internet application that enables easy capture and archival of information regarding our daily meals. The primary purpose of FoodLog is to facilitate dietary management support with minimum manual recording of information. It analyzes image archives that belong to a user to identify images of meals. Further image analysis determines the nutritional composition of these meals and stores the data to form a log. The user can view the data from this log in different formats, and also edit the data to correct any mistakes that occurred during image analysis. This application was recently opened to the public, and had accumulated approximately 25000 images during the first two months since its launch. We present the current status of this application, and discuss our future plans to extend it to allow interaction between users and more effective dietary management.","PeriodicalId":348792,"journal":{"name":"2010 16th International Conference on Virtual Systems and Multimedia","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134318638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-12-13DOI: 10.1109/VSMM.2010.5665941
Diego Roberto Colombo Dias, Anthony F. La Marca, Affonso Moia Vieira, Mário Popolin Neto, J. R. Ferreira Brega, M. Guimarães, José Roberto Pereira Lauris
This paper presents the development of an multi-projection stereoscopic dental arches application with semantic descriptions. The first section presents the concepts of the used technologies. Applications and examples are demonstrated. Finally, is presented the physical structure and the developed system, where a 3D dental arch is used as a model and can be viewed in multi-projection, thereby, providing greater user's immersion.
{"title":"Dental arches multi-projection system with semantic descriptions","authors":"Diego Roberto Colombo Dias, Anthony F. La Marca, Affonso Moia Vieira, Mário Popolin Neto, J. R. Ferreira Brega, M. Guimarães, José Roberto Pereira Lauris","doi":"10.1109/VSMM.2010.5665941","DOIUrl":"https://doi.org/10.1109/VSMM.2010.5665941","url":null,"abstract":"This paper presents the development of an multi-projection stereoscopic dental arches application with semantic descriptions. The first section presents the concepts of the used technologies. Applications and examples are demonstrated. Finally, is presented the physical structure and the developed system, where a 3D dental arch is used as a model and can be viewed in multi-projection, thereby, providing greater user's immersion.","PeriodicalId":348792,"journal":{"name":"2010 16th International Conference on Virtual Systems and Multimedia","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127190253","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-12-13DOI: 10.1109/VSMM.2010.5665974
Gowun Jeong, H. Yang
One of the general objectives of visual surveillance is to recognise abnormal activities from images. Current object detection/tracking techniques cannot directly classify such activities as fighting and snatching, while they reliably recognise primitive actions, such as walking and running. We represent each target activity as ground, weighted and undirected trees, Markov logic networks (MLNs), starting with primitive actions at the bottom and activities on top, using Horn clauses. The likelihood of one ground activity at root gives a reliable probability that the event actually happens. Computing such a probability could be intractable unless the truth values of all the nodes in a given network are known in advance. This study proposes two methods to infer such unknown values in exploitative and explorative manners. An additional modification of MLNs is also considered to improve accuracy of recognition. The experiments by means of unknown value inference methods and modification of MLNs present that these approaches overcome several well-known limitations that the conventional researches have experienced.
{"title":"Context-aware activity recognition by Markov logic networks of trained weights","authors":"Gowun Jeong, H. Yang","doi":"10.1109/VSMM.2010.5665974","DOIUrl":"https://doi.org/10.1109/VSMM.2010.5665974","url":null,"abstract":"One of the general objectives of visual surveillance is to recognise abnormal activities from images. Current object detection/tracking techniques cannot directly classify such activities as fighting and snatching, while they reliably recognise primitive actions, such as walking and running. We represent each target activity as ground, weighted and undirected trees, Markov logic networks (MLNs), starting with primitive actions at the bottom and activities on top, using Horn clauses. The likelihood of one ground activity at root gives a reliable probability that the event actually happens. Computing such a probability could be intractable unless the truth values of all the nodes in a given network are known in advance. This study proposes two methods to infer such unknown values in exploitative and explorative manners. An additional modification of MLNs is also considered to improve accuracy of recognition. The experiments by means of unknown value inference methods and modification of MLNs present that these approaches overcome several well-known limitations that the conventional researches have experienced.","PeriodicalId":348792,"journal":{"name":"2010 16th International Conference on Virtual Systems and Multimedia","volume":"133 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130962241","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-12-13DOI: 10.1109/VSMM.2010.5665937
Wun-Bin Yang, Min-Bin Chen, Y. Yen, Hung-Ming Cheng
Historical architecture, ancestors' important cultural assets, transmits conventional environment and skill experience in the society and cultural rogress. However, due to the long process of confirming, the limitation of budget and timing, the preservation of historical architecture are not efficiently executed preservation. Moreover, natural disaster, such as earthquake, fire and collapse, caused history architecture disappeared at one moment. 3D laser scanning technology is a new trend in the 21st century. With the development of 3D laser scanner, the data of 3-dimensional coordination could be conserved as historical architecture in a digital aspect for more complete preservation.
{"title":"An integrated 3D laser scanning technique for the digitization of historic buildings","authors":"Wun-Bin Yang, Min-Bin Chen, Y. Yen, Hung-Ming Cheng","doi":"10.1109/VSMM.2010.5665937","DOIUrl":"https://doi.org/10.1109/VSMM.2010.5665937","url":null,"abstract":"Historical architecture, ancestors' important cultural assets, transmits conventional environment and skill experience in the society and cultural rogress. However, due to the long process of confirming, the limitation of budget and timing, the preservation of historical architecture are not efficiently executed preservation. Moreover, natural disaster, such as earthquake, fire and collapse, caused history architecture disappeared at one moment. 3D laser scanning technology is a new trend in the 21st century. With the development of 3D laser scanner, the data of 3-dimensional coordination could be conserved as historical architecture in a digital aspect for more complete preservation.","PeriodicalId":348792,"journal":{"name":"2010 16th International Conference on Virtual Systems and Multimedia","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128789795","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-12-13DOI: 10.1109/VSMM.2010.5665936
D. Kera, Connor Graham
Cultural heritage in Singapore is a contested zone in which various interests, starting with tourism marketing campaigns and ending with identity building of a multicultural society, compete over the function and the definition of the collective and personal past. How to preserve memories and experiences in a city that is changing rapidly and how to reflect upon the changes? How to negotiate between the disappearing and forgotten past and the omnipresent future in Singapore? With a team of five students we conducted a series of design experiments and probes to discover novel ways for motivating people to take a heritage walk which can engage both locals and nonlocals into experiencing Singapore and its ethnic and cultural diversity. In our working prototype “Living Avatars Network” we evaluated a design idea of a real-time interface for outsourcing experiences as an incentive for a special type of walk which connects not only the past with the present but also serves as a platform for intercultural and intergenerational dialogue. Singapore is an ideal place to test such interactions between the emerging technologies and the disappearing traditions. In our project we designed a novel form of “practicing” and reliving cultural heritage in the age of ubiquitous and real time technologies that bring very different temporalities into play.
{"title":"Cultural heritage in the age of real time media: developing the Living Avatars Network","authors":"D. Kera, Connor Graham","doi":"10.1109/VSMM.2010.5665936","DOIUrl":"https://doi.org/10.1109/VSMM.2010.5665936","url":null,"abstract":"Cultural heritage in Singapore is a contested zone in which various interests, starting with tourism marketing campaigns and ending with identity building of a multicultural society, compete over the function and the definition of the collective and personal past. How to preserve memories and experiences in a city that is changing rapidly and how to reflect upon the changes? How to negotiate between the disappearing and forgotten past and the omnipresent future in Singapore? With a team of five students we conducted a series of design experiments and probes to discover novel ways for motivating people to take a heritage walk which can engage both locals and nonlocals into experiencing Singapore and its ethnic and cultural diversity. In our working prototype “Living Avatars Network” we evaluated a design idea of a real-time interface for outsourcing experiences as an incentive for a special type of walk which connects not only the past with the present but also serves as a platform for intercultural and intergenerational dialogue. Singapore is an ideal place to test such interactions between the emerging technologies and the disappearing traditions. In our project we designed a novel form of “practicing” and reliving cultural heritage in the age of ubiquitous and real time technologies that bring very different temporalities into play.","PeriodicalId":348792,"journal":{"name":"2010 16th International Conference on Virtual Systems and Multimedia","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123295317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-12-13DOI: 10.1109/VSMM.2010.5665955
J. Watanabe, Ryoko Ueoka
This paper describes novel perceptual experiences, which are achieved by media technologies, from the viewpoint of in-betweenness. In addition, we extended the viewpoint of analysis to the fields of communication and fashion.
本文从中间性的角度描述了媒介技术所实现的新型感知体验。此外，我们将分析的观点扩展到传播和时尚领域。
{"title":"Enhanced boundary of body image - perception & fashion","authors":"J. Watanabe, Ryoko Ueoka","doi":"10.1109/VSMM.2010.5665955","DOIUrl":"https://doi.org/10.1109/VSMM.2010.5665955","url":null,"abstract":"This paper describes novel perceptual experiences, which are achieved by media technologies, from the viewpoint of in-betweenness. In addition, we extended the viewpoint of analysis to the fields of communication and fashion.","PeriodicalId":348792,"journal":{"name":"2010 16th International Conference on Virtual Systems and Multimedia","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116612627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-12-13DOI: 10.1109/VSMM.2010.5665990
Hafizur Rahaman, M. Rashid, Masudur Rahman
The 8th century Buddhist Monastery of Sompur Mahaviahara at Paharpur drew deep attention by the architectural historians of the South Asia from the very discovery of the ruins for its unique architectural features and strategic spatio-temporal location. During the first phase of study, we have developed a preliminary reconstructed 3D model of Sompur Mahaviahara - as ‘professional interpretation’[1]. But this attempt was hindered due to non-availability of substantial resources. At the same time we have found it almost impossible to depend solely on materials that are available at first hand to demonstrate a continuous narrative of this monument. This leads us to a second phase of study which attempts to collect and investigate ‘public interpretation’ of the same monument. We propose a novel approach of heritage interpretation for general people, opposing the traditional interpretation method. We have developed an interactive web-portal, where active online participants can collaborate and collectively generate a knowledge base of cultural memories. We hope through this participatory approach of reconstruction/interpretation will provide an understanding of public view about this monument and at the same time will generate a popular database which can later support our previous study to get a clearer picture of the past glory of Sompur Mahavihara. As a methodology ‘heritage interpretation’ is first conceptualized; followed by elaboration of professional interpretation. Similar cases have been studied and some new strategies have been picked to develop an online portal.
{"title":"Heritage interpretation: Collective reconstruction of Sompur Mahavihara, Bangladesh","authors":"Hafizur Rahaman, M. Rashid, Masudur Rahman","doi":"10.1109/VSMM.2010.5665990","DOIUrl":"https://doi.org/10.1109/VSMM.2010.5665990","url":null,"abstract":"The 8th century Buddhist Monastery of Sompur Mahaviahara at Paharpur drew deep attention by the architectural historians of the South Asia from the very discovery of the ruins for its unique architectural features and strategic spatio-temporal location. During the first phase of study, we have developed a preliminary reconstructed 3D model of Sompur Mahaviahara - as ‘professional interpretation’[1]. But this attempt was hindered due to non-availability of substantial resources. At the same time we have found it almost impossible to depend solely on materials that are available at first hand to demonstrate a continuous narrative of this monument. This leads us to a second phase of study which attempts to collect and investigate ‘public interpretation’ of the same monument. We propose a novel approach of heritage interpretation for general people, opposing the traditional interpretation method. We have developed an interactive web-portal, where active online participants can collaborate and collectively generate a knowledge base of cultural memories. We hope through this participatory approach of reconstruction/interpretation will provide an understanding of public view about this monument and at the same time will generate a popular database which can later support our previous study to get a clearer picture of the past glory of Sompur Mahavihara. As a methodology ‘heritage interpretation’ is first conceptualized; followed by elaboration of professional interpretation. Similar cases have been studied and some new strategies have been picked to develop an online portal.","PeriodicalId":348792,"journal":{"name":"2010 16th International Conference on Virtual Systems and Multimedia","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132630852","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-12-13DOI: 10.1109/VSMM.2010.5665957
Ryota Ueno
This paper proposes the new way to exhibit digital cultural assets by using virtual reality system. We introduce our trial for promoting people to understand the cultural assets at the Tokyo National Museum. And by evaluating the result of questionnaires, we verify the validity of our methods.
{"title":"An experience of digital cultural assets at museum","authors":"Ryota Ueno","doi":"10.1109/VSMM.2010.5665957","DOIUrl":"https://doi.org/10.1109/VSMM.2010.5665957","url":null,"abstract":"This paper proposes the new way to exhibit digital cultural assets by using virtual reality system. We introduce our trial for promoting people to understand the cultural assets at the Tokyo National Museum. And by evaluating the result of questionnaires, we verify the validity of our methods.","PeriodicalId":348792,"journal":{"name":"2010 16th International Conference on Virtual Systems and Multimedia","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115180739","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：