{"title":"Data-Driven Hair Modeling from a Single Image","authors":"Jiqiang Wu, Yongtang Bao, Yue Qi","doi":"10.1109/ICVRV.2018.00010","DOIUrl":"https://doi.org/10.1109/ICVRV.2018.00010","url":null,"abstract":"Hair is one of the most distinctive part of a human body, which is essential for the digitization of compelling virtual avatars. We present a data-driven approach for generating complete and complex 3D hairstyles from a single-view portrait. We construct a hairstyle database which contains more than 2000 hair models. Given a target hairstyle portrait image, we first draw a few strokes as guidance. We then search multiple best matching examples from the database based on our enhanced matching algorithm. Finally, we combine them consistently into a single hairstyle. The generated hairstyles are visually comparable to original portrait images. The reconstructed 3D hair models can be used for many applications, such as hair editing, and dynamic hair simulation.","PeriodicalId":159517,"journal":{"name":"2018 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115587390","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":"Traffic Signs Detection and Augmented Reality Based on Multithreading","authors":"Wenting Li, Qian Li, Shangbing Gao, C. Cai","doi":"10.1109/icvrv.2018.00025","DOIUrl":"https://doi.org/10.1109/icvrv.2018.00025","url":null,"abstract":"The detection algorithm of existing traffic signs was improved, and the implementation of traffic sign detection and recognition was described in detail. This article mainly contains the following: Through targeted feature extraction, a trained cascaded classifier is used to obtain the location of traffic signs. Combined with the previous detection results, feature extraction is performed on the specified part of the target. The results are analyzed according to the location of detection or the results of the trained support vector machine are used to classify, and the classification results are obtained to realize the recognition of traffic signs. The total number of samples used for goal training in this article has reached thousands. The test results show that the detection rate of the traffic sign under the trained scene has been more than 90%. The algorithm has been optimized by the multi-thread method and combined with augmented reality to achieve real-time feedback.","PeriodicalId":159517,"journal":{"name":"2018 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123139982","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":"[Title page i]","authors":"","doi":"10.1109/icvrv.2018.00001","DOIUrl":"https://doi.org/10.1109/icvrv.2018.00001","url":null,"abstract":"","PeriodicalId":159517,"journal":{"name":"2018 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"46 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134599636","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}
Ji Wang, Tianlu Mao, Xiyuan Song, Shaohua Liu, Hao Jiang, Zhaoqi Wang
{"title":"Parallel Crowd Simulation Based on Power Law","authors":"Ji Wang, Tianlu Mao, Xiyuan Song, Shaohua Liu, Hao Jiang, Zhaoqi Wang","doi":"10.1109/ICVRV.2018.00023","DOIUrl":"https://doi.org/10.1109/ICVRV.2018.00023","url":null,"abstract":"Crowd simulation technology is increasingly used in the fields of film, animation, games, military training, and public safety. In this paper, we propose a parallel crowd simulation model based on Power Law. This model use CUDA architecture to parallelize the Power Law model on GPU, so that each agent's behavior simulation is synchronized on different threads. This model considers the fine effects of the microscopic model while significantly improving the model simulation efficiency, making it possible to simulate large-scale crowds in real time and can accurately obtain the status information of each agent in each frame.","PeriodicalId":159517,"journal":{"name":"2018 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127500347","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":"Diabetic Retinopathy Detection Based on Deep Convolutional Neural Networks for Localization of Discriminative Regions","authors":"Junjun Pan, Yong Zhifan, Sui Dong, Qin Hong","doi":"10.1109/icvrv.2018.00016","DOIUrl":"https://doi.org/10.1109/icvrv.2018.00016","url":null,"abstract":"Diabetic Retinopathy (DR) is the leading cause of avoidable vision impairment. Currently, manual DR detection is a time consuming task, which relies on well-trained clinicians with skills. In this paper, we propose a novel and automatic diabetic retinopathy (DR) detection method using deep convolutional neural networks (DCNNs). To identify the region of interests (ROIs), we design an attention mechanism for scoring the specific regions, refered as regions scoring map (RSM). The RSM is based on deep convolutional neural networks, which are trained only with image-level labels on a large scale DR dataset. Specifically, the RSM is mainly inserted into deep residual networks between intermediate stages. With RSM, the proposed model can score the different regions of an retina image to highlight the discriminative ROIs in terms of image severity level. In experiments, around 30000 colour retinal images are used to train the proposed model and around 5000 images are collected to evaluate its classification performance. The results show that our DCNN model can obtain comparable performance while achieving the merits of providing the RSM to locate the discriminative regions of the input image.","PeriodicalId":159517,"journal":{"name":"2018 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130619804","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":"Distributed Refinement of Large-Scale 3D Mesh for Accurate Multi-View Reconstruction","authors":"Qing Luo, Yao Li, Yue Qi","doi":"10.1109/ICVRV.2018.00018","DOIUrl":"https://doi.org/10.1109/ICVRV.2018.00018","url":null,"abstract":"As the scene of multi-view reconstruction becomes larger, a single machine can no longer satisfy the refinement of 3D mesh in large scenes including mesh simplification, subdivision, smoothness and recovering meaningful details. In this paper, We propose a distributed method to refine a large-scale 3D mesh for accurate multiview reconstruction. First, we divide the initial mesh into blocks directly, which can utilize computing power of each computer. And then we make simplification and subdivision on those blocks, which can reduce mesh's noise and remove redundant vertices, so as to generate a high quality mesh where the difference of the size of each edge is not too large. Next, we propose to split a graph consisting of multiple images in order to minimize the overlapped image data in each block. Finally, we use distributed variational surface refinement algorithm to capture meaningful details of mesh. The experiments on both public large scale data-sets and our very large scale aerial photo sets demonstrate that the proposed distributed method is fast and robust, and is suitable for all kinds of large scene areas.","PeriodicalId":159517,"journal":{"name":"2018 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128195059","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}