{"title":"Multiple Features Fusion Based Inverted Multi-index for Image Retrieval","authors":"Xiangbin Shi, Zhongqiang Guo, Deyuan Zhang, Xuejian Fang","doi":"10.1109/ICVRV.2015.39","DOIUrl":"https://doi.org/10.1109/ICVRV.2015.39","url":null,"abstract":"In the community of image retrieval, single feature shows a very low discriminative power for matching, and typically the frequently-used SIFT feature only describes the local gradient distribution. Therefore, false matches occur prevalently. Besides, conventional inverted index usually returns long candidate lists for queries, with sparse subdivision of search space and limited accuracy. To tackle these problems, this paper proposes an effective image retrieval framework: multiple features fusion based inverted multi-index (mFFMI). Inside this framework, features are incorporated into inverted multi-index, providing respective complementary image information for accurate image retrieval. We fuse Root SIFT (a variant of SIFT), color and texture features into mFFMI framework, and experiments on benchmark datasets show that mFFMI structure yields favorable accuracy improvement, compared with state-of-the-art. Moreover, another significant characteristic of mFFMI is that it is well compatible with many other techniques, and a better performance can be achieved if combined with such prior techniques.","PeriodicalId":376143,"journal":{"name":"2015 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"141 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123774121","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":"BVCNN: A Multi-object Image Recognition Method Based on the Convolutional Neural Networks","authors":"Hui-ling Shi, Xiaodong Mu, Shuyang Wang","doi":"10.1109/ICVRV.2015.28","DOIUrl":"https://doi.org/10.1109/ICVRV.2015.28","url":null,"abstract":"This article puts forward a kind of huge amounts of multi-object image recognition method -- BVCNN. Firstly, BING method is used to recognize images, which greatly reduces the time of estimating image targets, and makes it possible that quickly identify multiple target images, compared to traditional convolution neural networks only achieving single target image recognition, Secondly, vectorization of deep convolutional neural networks is used for deep learning of characteristics in local image and recognition, which speeds up network training and testing, thirdly, using the context information in multi-object image classification, to a certain extent, helps to distinguish individual of similar characteristics according to environment, improving the multi-object image recognition accuracy. According to experiments, identifying a single image by this model only need less than 1 s, and this model can be used for image information fusion.","PeriodicalId":376143,"journal":{"name":"2015 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126651452","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":"An Image Search Method for Chinese Character Inscription Based on Optical Flow","authors":"Qing Zuo, Yue Qi","doi":"10.1109/ICVRV.2015.21","DOIUrl":"https://doi.org/10.1109/ICVRV.2015.21","url":null,"abstract":"Chinese character inscription is an important part of Chinese culture. With China's increasing efforts to protect traditional culture, more and more Chinese character inscription images are captured and stored in the digital culture museum. How to search a Chinese character is becoming an interesting research topic. The most important thing for searching an image is to define the similarity between two images. In this paper, we propose an image search method for Chinese character inscription images based on optical flow method. The two images are aligned with the optical flow and then the distance is measured. Considering with the noise in the images, a novel optical flow model with a L1-regularization for data term and a L2 smoothness term is proposed. We have tested our method on a database with more than one thousand images. And our method can return good results in several seconds.","PeriodicalId":376143,"journal":{"name":"2015 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128639978","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}
Changhua Yu, Min Xiong, Wei Fang, Quan Xu, Peng Zheng, Xianhong Zhang, Juelin Leng
{"title":"Research on the Algorithm for Grid-Based Hexahedral Mesh Generation of CAD Assembly Models","authors":"Changhua Yu, Min Xiong, Wei Fang, Quan Xu, Peng Zheng, Xianhong Zhang, Juelin Leng","doi":"10.1109/ICVRV.2015.12","DOIUrl":"https://doi.org/10.1109/ICVRV.2015.12","url":null,"abstract":"According to the geometric sharp features of CAD assembly models, an algorithm for the automated generation of grid-based full hexahedral element meshes was developed. This method involves six steps: geometric feature identification, cubic grid structure generation, jagged core mesh generation, surface matching, topological optimization and node smoothing. In the grid-based hexahedral mesh, the quality of the inner mesh is fine, the quality and the topological connections are very bad. Considering the specialty of the grid-based, this paper focused on the research for the surface matching and the topological optimization techniques in the grid-based hexahedral mesh generation. Among the surface matching process in this paper, a new matching algorithm combining the embedding technique was proposed, establishing the corresponding the boundary element relation between solid models and core grid, to make the hexahedral mesh accurately describe the geometric features of multiple-component assemblies. Among the topological optimization process, a new optimization technique was proposed, building the hexahedral shrink sets for the every geometry body and geometry surface, and then eliminating the hexahedral elements with bad topological connection by adding a layer of hexahedral elements on the boundary of the corresponding shrink set. Numerical examples show that the hexahedral mesh generated by the algorithm proposed in this paper described accurately the geometric features of the numerical models, and eliminated the poor hexahedral elements.","PeriodicalId":376143,"journal":{"name":"2015 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"125 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124609405","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":"A 2D Engineering Drawing and 3D Model Matching Algorithm for Process Plant","authors":"Rui Wen, Weiqing Tang, Zhiyong Su","doi":"10.1109/ICVRV.2015.13","DOIUrl":"https://doi.org/10.1109/ICVRV.2015.13","url":null,"abstract":"Process plant Computer-Aided Design (CAD) models can be roughly divided into 3D models and 2D engineering drawings. Matching calculation of these CAD models not only contributes to their coherency verification, but also benefits for the development of model retrieval. However, in process plant area, 3D models and 2D engineering drawings are different in both graphical representations and content structures, which leads to the inapplicability of current shape-feature based 2D & 3D matching approaches. To resolve this problem, a topological structure based algorithm is proposed. Exploiting component as the basic unit, we firstly transform 2D engineering drawings and 3D models into attribute graphs. Then, by introducing related researches of graph similarity, we calculate attribute graph similarity to measure the matching degree between their corresponding CAD models. The proposed algorithm is translation, rotation and similarity transformation consistent. Experimental results demonstrate its effectiveness and feasibility.","PeriodicalId":376143,"journal":{"name":"2015 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134080997","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":"Hand Posture Recognition Using Kinect","authors":"Jinhong Xie, Xukun Shen","doi":"10.1109/ICVRV.2015.9","DOIUrl":"https://doi.org/10.1109/ICVRV.2015.9","url":null,"abstract":"In this paper, we introduce a two-stage Hand Posture Recognition system using a Kinect sensor. In the first stage, we propose an effective algorithm to segment hand region from complex background, which incorporates both color, depth and skeleton information, without specific requirements on uniform-colored or stable background. It is robust with skin-color noise like arms and face, and can handle both-hands gesture. In the second stage, we use a standard Nearest Neighbor to recognize the posture. Experiments shows that the proposed system achieves high recognition accuracy.","PeriodicalId":376143,"journal":{"name":"2015 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"129 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134202137","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":"Remote Sensing Image Classification Method Based on Superpixel Segmentation and Adaptive Weighting K-Means","authors":"L. Ke, You Xiong, Wan Gang","doi":"10.1109/ICVRV.2015.35","DOIUrl":"https://doi.org/10.1109/ICVRV.2015.35","url":null,"abstract":"To improve remote sensing image classification precision, we propose a novel method which is based on super pixel and adaptive weighted K-Means. First, super pixel segmentation algorithm is used to divide input images into irregular blocks which remain their semantic information and boundaries. And then, SIFT, GIST, Census, Gabor, and Color histogram, and many other types of features are extracted. These five features represent different kinds of image characteristics. In order to obtain best feature combination, abundant experiments and analyses are performed. We also propose a novel adaptive weighted k-Means method to automatically estimate optimal weights and cluster centers for improving the representation accuracy of visual bag-of-words. An improved soft vector quantization based on sparse coding is adopted to generate the image feature. Finally support vector machine is utilized to complete the image classification. Experiments show that the new method proposed in this paper can effectively improve the classification accuracy of remote sensing images, and it also show good stability and robustness.","PeriodicalId":376143,"journal":{"name":"2015 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127426273","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":"Research on Combination Positioning Based on Natural Features and Gyroscopes for AR on Mobile Phones","authors":"Libo Sun, Jie Du, Wenhu Qin","doi":"10.1109/ICVRV.2015.55","DOIUrl":"https://doi.org/10.1109/ICVRV.2015.55","url":null,"abstract":"Considering the situation that performing indoor positioning system on mobile platform mostly relying on additional wireless signal and known environmental information, we propose an autonomous combination positioning method aimed at continuously estimating metric position and pose in an unknown scene. The vision-based SLAM algorithm extracts the image characteristics of the environment for producing a 3D map of point features and tracking erratic hand-held motion robustly. The position and pose calculated by SLAM algorithm are viewed as observation of an extended Kalman filter to correct the pose obtained from SINS. The initial value of metric scale of position is calculated using Least Squares which ensures the fast convergence of EKF. SINS is also used to detect the failures of vision part and improve robustness of tracking results in the environment with rare characteristics. The experimental results demonstrate that our system can produce detailed maps with thousands of landmarks which can be tracked at about 30HZ with high accuracy and robustness, although the computational resource is limited on current smart phones.","PeriodicalId":376143,"journal":{"name":"2015 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117156021","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":"Real-Time 3D Video Acquisition and Auto-Stereoscopic Display End-to-End Algorithm Based on Tiled Multi-projectors","authors":"Huayuan Guo, K. Qin, Feng Sun","doi":"10.1109/ICVRV.2015.51","DOIUrl":"https://doi.org/10.1109/ICVRV.2015.51","url":null,"abstract":"In order to display the real-time 3D video with more stereoscopic vision, this paper presents a novel end-to-end auto-stereoscopic tiled-projection display algorithm including dual-view video streaming acquisition, compression, transmission, and visualization. On the server side, it gets 3D video captured by dual-camera or read from dual-view 3D video files. Then, the dual-view images are transcoded into MPEG-2 /H.264 data by the Intel Quick Sync Video engine, and distributed to all rendering clients through the real-time transport protocol. On the client side, a self-calibrated method based on a slope voting policy is designed to eliminate the dual-viewpoint perspective distortion. After performing the geometric calibration and luminance correction, the dual-view images are interleaved on the optical display screen with a two-dimensional projection resolution of 3584 × 1536 pixels. Experimental results have validated that the projection system can bring viewers much better glasses-free stereoscopic vision at above 21 × 2 frames per second.","PeriodicalId":376143,"journal":{"name":"2015 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128235733","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}
Hongjuan Bu, Qing Xu, Shang Wu, Yuejun Guo, M. Sbert
{"title":"Detection and Removal for Impulse Noise in Monte Carlo Global Illumination Rendered Images of Highly Glossy Scenes","authors":"Hongjuan Bu, Qing Xu, Shang Wu, Yuejun Guo, M. Sbert","doi":"10.1109/ICVRV.2015.24","DOIUrl":"https://doi.org/10.1109/ICVRV.2015.24","url":null,"abstract":"Monte Carlo global illumination (MCGI) plays an important role for the rendering and synthesis of photorealistic image. However, the synthetic images using MCGI are always corrupted by heavy noise. Especially, very unpleasant impulse noise appears in MCGI images when 3D environments contain highly glossy surfaces. In this paper, we propose a two-stage filtering technique, by utilizing pixel samples, to remove the impulse noise during MCGI image rendering. Firstly, impulsive pixels are detected using the impulsiveness of the distribution. And then the noise suppression technique is applied to obtain the final result. For a salt-like noise, the outlier samples are identified by the Grubbs test and their energies are re-distributed to the neighbors of the noisy pixel. For a pepper-like noise, the energies of its neighbors are gathered. Experimental results show that the proposed method behaves remarkably well in terms of both quantitative assessment of noise reduction and qualitative evaluation of image quality.","PeriodicalId":376143,"journal":{"name":"2015 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127684760","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}