{"title":"High Dimensional Latent Space Variational AutoEncoders for Fake News Detection","authors":"Saad Sadiq, Nicolas Wagner, M. Shyu, D. Feaster","doi":"10.1109/MIPR.2019.00088","DOIUrl":"https://doi.org/10.1109/MIPR.2019.00088","url":null,"abstract":"With the advent of social media and cell phones, news is now far more reaching and impactful than ever before. This comes with the exponential increase in fake news that blurs the lines of reality and holds the power to sway public opinion. To counter the impact of fake news, several research groups have developed novel algorithms that could fact check news as a human would do. Unfortunately, natural language processing (NLP) is a complicated task because of the underlying hidden meanings in human communication. In this paper, we propose a novel method that builds a latent representation of natural language to capture its underlying hidden meanings accurately and classify fake news. Our approach connects the high-level semantic concepts in the news content with their low-level deep representations so that the complex news text consisting of satire, sarcasm, and purposeful misleading content can be translated into quantifiable latent spaces. This allows us to achieve very high accuracy, surpassing the scores of all winners of the fake news challenge.","PeriodicalId":201969,"journal":{"name":"2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133080574","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}
Mahmut Unan, Junmo An, I. Seimenis, D. Shah, N. Tsekos
{"title":"3D Reconstruction of Tubular Structure Using Radially Deployed Projections","authors":"Mahmut Unan, Junmo An, I. Seimenis, D. Shah, N. Tsekos","doi":"10.1109/MIPR.2019.00064","DOIUrl":"https://doi.org/10.1109/MIPR.2019.00064","url":null,"abstract":"Acquiring volumetric data plays a crucial role in the field of medical imaging. 3D reconstruction is mostly performed using multislice image datasets. The objective of this research is to introduce a magnetic resonance technique for imaging tubular structures and their 3D reconstructions using multiple radially deployed projections. The oblique projection sequence was evaluated on a phantom, and multislice dataset is collected using the same phantom for the reference. To compute the correctness of the 3D reconstruction process, the resulting meshes were compared using the Hausdorff Distance Calculation and Point Cloud Comparison methods.","PeriodicalId":201969,"journal":{"name":"2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130090721","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}
Hanchao Yu, Yang Fu, Haichao Yu, Yunchao Wei, Xinchao Wang, Jianbo Jiao, Matthew Bramler, T. Kesavadas, Humphrey Shi, Zhangyang Wang, B. Wen, Thomas S. Huang
{"title":"A Novel Framework for 3D-2D Vertebra Matching","authors":"Hanchao Yu, Yang Fu, Haichao Yu, Yunchao Wei, Xinchao Wang, Jianbo Jiao, Matthew Bramler, T. Kesavadas, Humphrey Shi, Zhangyang Wang, B. Wen, Thomas S. Huang","doi":"10.1109/MIPR.2019.00029","DOIUrl":"https://doi.org/10.1109/MIPR.2019.00029","url":null,"abstract":"3D-2D medical image matching is a crucial task in image-guided surgery, image-guided radiation therapy and minimally invasive surgery. The task relies on identifying the correspondence between a 2D reference image and the 2D projection of 3D target image. In this paper, we propose a novel image matching framework between 3D CT projection and 2D X-ray image, tailored for vertebra images. The main idea is to learn a vertebra detector by means of deep neural network. The detected vertebra is represented by a bounding box in the 3D CT projection. Next, the bounding box annotated by the doctor on the X-ray image is matched to the corresponding box in the 3D projection. We evaluate our proposed method on our own-collected 3D-2D registration dataset. The experimental results show that our framework outperforms the state-of-the-art neural network-based keypoint matching methods.","PeriodicalId":201969,"journal":{"name":"2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132388869","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}
Yuki Kanaya, S. Kawanaka, H. Suwa, Yutaka Arakawa, K. Yasumoto
{"title":"Automatic Tour Video Summarization Focusing on Scene Change for Advance Touristic Experience","authors":"Yuki Kanaya, S. Kawanaka, H. Suwa, Yutaka Arakawa, K. Yasumoto","doi":"10.1109/MIPR.2019.00079","DOIUrl":"https://doi.org/10.1109/MIPR.2019.00079","url":null,"abstract":"In this paper, we propose a novel time-lapse based video summarization method without substantial loss of information important for viewers to plan a tour route. In the proposed method, we focus on scene changes in the video. We extract scenes with a certain level of changes compared to previous scenes as important (slowly played back) in the summarized video, while other scenes are fast forwarded. To evaluate the effectiveness of our proposed method, comparing the video summarized with our method and that summarized manually (by voting for necessary/unnecessary scenes), our method identified the important scenes with an F-measure of 62.22%.","PeriodicalId":201969,"journal":{"name":"2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129224390","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}
{"title":"Stalker Retrieval on Surveillance Videos Using Spatio-Temporal Coappearance","authors":"Jianquan Liu, Duncan Yung, Shoji Nishimura, Takuya Araki","doi":"10.1109/MIPR.2019.00030","DOIUrl":"https://doi.org/10.1109/MIPR.2019.00030","url":null,"abstract":"Stalking is a distinctive form of criminal activity that consists of repeated following, watching, tracking and/or harassing by one person against another. Stalking victims are reported many daily happenings, thus stalking has attracted widespread public concern in recent years. It results in that stalking analysis becomes a significant social problem in criminology. Meanwhile, nowadays surveillance camera has become a ubiquitous aspect of the modern urban landscape, widely installed for security purposes. Video data generated by surveillance cameras are recognized as reliable sources for crime prevention and investigation, definitely useful for stalking analysis. However, since stalking behaviors have complicated variants, it is too difficult to directly analyze stalking behaviors in surveillance videos in terms of behavior analysis. In the state of the art, even there is no existing technology that can handle such analysis. Hence, motivated by this and inspired by the distinctive characteristics of stalking behaviors, in this paper, we define a new query type, stalker query, to retrieve potential stalkers. We then propose efficient algorithms to handle the stalker query, and conduct extensive experiments based on real surveillance videos to evaluate the efficiency and scalability of our algorithms. The experimental results show that our approach outperforms the baseline method over two orders of magnitude to return stalker candidates. Besides, we also examine the high accuracy of retrieval by a data set containing simulated real stalkers. This data set will be contributed to academic communities for attracting continuous effort to tackle remaining open problems with high social impact.","PeriodicalId":201969,"journal":{"name":"2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115315093","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}
{"title":"Efficient Incremental Training for Deep Convolutional Neural Networks","authors":"Yudong Tao, Y. Tu, M. Shyu","doi":"10.1109/MIPR.2019.00058","DOIUrl":"https://doi.org/10.1109/MIPR.2019.00058","url":null,"abstract":"While the deep convolutional neural networks (DCNNs) have shown excellent performance in various applications, such as image classification, training a DCNN model from scratch is computationally expensive and time consuming. In recent years, a lot of studies have been done to accelerate the training of DCNNs, but most of them were performed in a one-time manner. Considering the learning patterns of the human beings, people typically feel more comfortable to learn things in an incremental way and may be overwhelmed when absorbing a large amount of new information at once. Therefore, we demonstrate a new training schema that splits the whole training process into several sub-training steps. In this study, we propose an efficient DCNN training framework where we learn the new classes of concepts incrementally. The experiments are conducted on CIFAR-100 with VGG-19 as the backbone network. Our proposed framework demonstrates a comparable accuracy compared with the model trained from scratch and has shown 1.42x faster training speed.","PeriodicalId":201969,"journal":{"name":"2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122141439","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}
{"title":"When Green Screen Meets Panoramic Videos: An Interesting Video Combination Framework","authors":"Chenxi Feng, Tianyi Feng, Long Ye","doi":"10.1109/MIPR.2019.00097","DOIUrl":"https://doi.org/10.1109/MIPR.2019.00097","url":null,"abstract":"In this paper, an interesting video combination framework was established based on Unreal Engine 4. By substituting the green screen image with panorama videos, our framework could produce lifelike synthetic videos in which the foreground is the portrait captured by real-time TV cameras whereas the background is a pseudo scenario generated by previously captured panorama videos. The pseudo background updates itself from the panorama video according to the pose information of the TV camera so that the portrait looks as if immersed into the pseudo scenario rather than the green screen.","PeriodicalId":201969,"journal":{"name":"2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131523715","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}
{"title":"Using Amino Acids of Images for Identifying Pornographic Images","authors":"Vijay Naidu, A. Narayanan, M. Mohanty","doi":"10.1109/MIPR.2019.00010","DOIUrl":"https://doi.org/10.1109/MIPR.2019.00010","url":null,"abstract":"The pornographic images need to be regulated as they can have an adverse effect on the society. This paper purposes an image amino acid-based method to identify a pornographic image. The proposed method works by converting an image into biological amino acid form (e.g., to A, C, D, G, H, etc. form) and then extracting a signature from the image (which is represented in amino acid form) using biological tools. Using this method, a number of signatures are obtained from a set of few known pornographic images. These signatures are then matched against a database of images to find out the pornographic images. The matching is done using the openly available anti-virus scanner Clamscan. Here, the signatures obtained from the pornographic images are represented as signatures of viruses. The experimental result shows that the proposed method can identify a pornographic image with a high detection rate.","PeriodicalId":201969,"journal":{"name":"2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129779298","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}
{"title":"SecureCSearch: Secure Searching in PDF Over Untrusted Cloud Servers","authors":"Meet Shah, M. Mohanty, P. Atrey","doi":"10.1109/MIPR.2019.00069","DOIUrl":"https://doi.org/10.1109/MIPR.2019.00069","url":null,"abstract":"The usage of cloud for data storage has become ubiquitous. To prevent data leakage and hacks, it is common to encrypt the data (e.g. PDF files) before sending it to a cloud. However, this limits the search for specific files containing certain keywords over an encrypted cloud data. The traditional method is to take down all files from a cloud, store them locally, decrypt and then search over them, defeating the purpose of using a cloud. In this paper, we propose a method, called SecureCSearch, to perform keyword search operations on the encrypted PDF files over cloud in an efficient manner. The proposed method makes use of Shamir's Secret Sharing scheme in a novel way to create encrypted shares of the PDF file and the keyword to search. We show that the proposed method maintains the security of the data and incurs minimal computation cost.","PeriodicalId":201969,"journal":{"name":"2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128088914","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}
{"title":"Design and Implementation of Virtual-Real Interactive System for Mixed Reality","authors":"Xinxin Shi, Xin Xiang, Long Ye","doi":"10.1109/MIPR.2019.00096","DOIUrl":"https://doi.org/10.1109/MIPR.2019.00096","url":null,"abstract":"Mixed reality (MR) combines real and virtual worlds to create an interactive bridge between the real world, the virtual world and users. In order to make the interaction between the user and the virtual world intelligent, a virtual-real interaction system for mixed reality scheme is proposed. This paper mainly discusses the technical solutions for implementing the system and demonstrates the interaction between the user and virtual characters in the interactive system. The virtual characters of the interactive system are stimulated by real and virtual scenes, and the behavior of real people changes [1]. The modules used in this study include: 1) Unreal Engine 4 builds virtual scenes and characters, 2) Motion capture system captures real-world information, aligns real and virtual worlds. Alignment refers to the same position information and orientation information of the real world and the virtual world. 3) Graphics workstation synthesizes scenes captured by optical cameras and virtual scenes.","PeriodicalId":201969,"journal":{"name":"2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)","volume":"298 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133741116","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}