Pub Date : 2017-10-01DOI: 10.1109/ICVRV.2017.00011
Hexi Li, Na Jiang, Chenxin Sun, Zhong Zhou, Wei Wu
Multi-target tracking is a worthy studying issue in computer vision. For surveillance video, frequent occlusion and dense crowds complicate the issue. To resolve these difficulties, this paper proposes an effective algorithm of multi-target tracking in videos. Firstly, the faster Rcnn is proposed with the residual network to extract the objects of pedestrians in surveillance videos. The proposedment can effectively eliminate invalid target detection frames, separate peer targets and resist partial occlusions. Then, this paper put forward an accurate and efficient appearance-feature matching network model that is inspired by pedestrian re-identification theory. The deep learning feature-extraction module is composed of the stem Cnn and the Resnet blocks, therefore it can load res-50 caffemodel as pretraining model to increase the accuracy of the featureextraction. Meanwhile, the proposed network can decrease the time of train and test comparing with Resnet. Finally, the obtained multiple target tracking trajectories are further optimized by the strategy of occlusion distinction, deduplication and merging. The experiment results of the 2D MOT 2015 benchmark, KITTI dataset indicate that this proposed algorithm outperforms alternative multiple objects trackers in terms of multiple indicators.
多目标跟踪是计算机视觉中一个值得研究的问题。对于监控视频来说，频繁的遮挡和密集的人群使问题复杂化。为了解决这些问题，本文提出了一种有效的视频多目标跟踪算法。首先，利用残差网络提出了一种更快的Rcnn来提取监控视频中的行人目标。该方法可以有效地消除无效目标检测帧，分离对等目标，抵抗部分遮挡。然后，受行人再识别理论的启发，提出了一种准确高效的外观特征匹配网络模型。深度学习特征提取模块由stem Cnn和Resnet块组成，因此可以加载res-50 caffmodel作为预训练模型，以提高特征提取的准确性。同时，与Resnet相比，该网络减少了训练和测试的时间。最后，通过遮挡区分、重复数据删除和合并策略对得到的多目标跟踪轨迹进行进一步优化。2D MOT 2015基准KITTI数据集的实验结果表明，该算法在多指标方面优于备选多目标跟踪器。
{"title":"Learning Deep Appearance Feature for Multi-target Tracking","authors":"Hexi Li, Na Jiang, Chenxin Sun, Zhong Zhou, Wei Wu","doi":"10.1109/ICVRV.2017.00011","DOIUrl":"https://doi.org/10.1109/ICVRV.2017.00011","url":null,"abstract":"Multi-target tracking is a worthy studying issue in computer vision. For surveillance video, frequent occlusion and dense crowds complicate the issue. To resolve these difficulties, this paper proposes an effective algorithm of multi-target tracking in videos. Firstly, the faster Rcnn is proposed with the residual network to extract the objects of pedestrians in surveillance videos. The proposedment can effectively eliminate invalid target detection frames, separate peer targets and resist partial occlusions. Then, this paper put forward an accurate and efficient appearance-feature matching network model that is inspired by pedestrian re-identification theory. The deep learning feature-extraction module is composed of the stem Cnn and the Resnet blocks, therefore it can load res-50 caffemodel as pretraining model to increase the accuracy of the featureextraction. Meanwhile, the proposed network can decrease the time of train and test comparing with Resnet. Finally, the obtained multiple target tracking trajectories are further optimized by the strategy of occlusion distinction, deduplication and merging. The experiment results of the 2D MOT 2015 benchmark, KITTI dataset indicate that this proposed algorithm outperforms alternative multiple objects trackers in terms of multiple indicators.","PeriodicalId":187934,"journal":{"name":"2017 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115551133","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 : 2017-10-01DOI: 10.1109/ICVRV.2017.00043
Zhiyong Tu, Jinyuan Jia, Feng-ting Yan
Practical 3D printing application always encounters motor motion confliction and warping of ugly product. This paper presents the proper ways to solve these problems. Firstly, to be assured precise motion controlling, safety and 3D printing process stabilization, we present motion mutex approach with Karnaugh map for real time monitoring motor motion status, and its algorithm of motion mutex, based on this way to make motor motion coordination working to accomplish precise process controlling. Secondly, to obtain reliable and precise product with smooth surface and to reduce the time of fabrication product, we propose comprehensive path planning to achieve this purpose, which integrated contour offset scanning and short linear scanning. Finally, with these methods in the practice we have produced the beautiful statuette of warrior.
{"title":"A Reliable, Precise and Efficient 3D Printer Process Controlling Algorithm","authors":"Zhiyong Tu, Jinyuan Jia, Feng-ting Yan","doi":"10.1109/ICVRV.2017.00043","DOIUrl":"https://doi.org/10.1109/ICVRV.2017.00043","url":null,"abstract":"Practical 3D printing application always encounters motor motion confliction and warping of ugly product. This paper presents the proper ways to solve these problems. Firstly, to be assured precise motion controlling, safety and 3D printing process stabilization, we present motion mutex approach with Karnaugh map for real time monitoring motor motion status, and its algorithm of motion mutex, based on this way to make motor motion coordination working to accomplish precise process controlling. Secondly, to obtain reliable and precise product with smooth surface and to reduce the time of fabrication product, we propose comprehensive path planning to achieve this purpose, which integrated contour offset scanning and short linear scanning. Finally, with these methods in the practice we have produced the beautiful statuette of warrior.","PeriodicalId":187934,"journal":{"name":"2017 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121017266","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 : 2017-10-01DOI: 10.1109/ICVRV.2017.00047
Dazhi Zhan, Xiangrong Zeng, Weili Li, Yu Liu, Z. Xiong
A method for image quality enhancement of simple lens image under low-illumination condition is presented in this paper. We first introduce an accurate camera-scene alignment framework that generates exactly matched and tone-consistent image pairs to estimate blur kernel. We then use the cross-channel matrix obtained in the tone curve calibration to restore the details of the blurred image. Finally, the clear image is restored by non-blind deconvolution and compared with other methods to prove the advantages of our method.
{"title":"Low-Light Image Deblurring Based on Simple Lens System","authors":"Dazhi Zhan, Xiangrong Zeng, Weili Li, Yu Liu, Z. Xiong","doi":"10.1109/ICVRV.2017.00047","DOIUrl":"https://doi.org/10.1109/ICVRV.2017.00047","url":null,"abstract":"A method for image quality enhancement of simple lens image under low-illumination condition is presented in this paper. We first introduce an accurate camera-scene alignment framework that generates exactly matched and tone-consistent image pairs to estimate blur kernel. We then use the cross-channel matrix obtained in the tone curve calibration to restore the details of the blurred image. Finally, the clear image is restored by non-blind deconvolution and compared with other methods to prove the advantages of our method.","PeriodicalId":187934,"journal":{"name":"2017 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132759085","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 : 2017-10-01DOI: 10.1109/ICVRV.2017.00080
Xiang Nan, Zhou Zehong, Pan Zhigeng
Generating emotion contagion results were very important in crowd simulation field. However, as the individuals often moving dynamically, then computing the contagion process becomes a challenge. In this paper, we focused on the emotion contagion effects on dynamic aggregation process of the virtual pedestrian. Firstly, according to the social force theory, we constructed individuals' moving velocities based on their expectations. Secondly, made an adjacent test to generate the nearer neighbors as emotional contagions usually occurred between neighbors. And then we treated the emotional contagions between individuals and their neighbors as the information spreading process so that we calculated the influences on their moving directions through the emotional information spreading model SIR (Susceptible Infective Removal). Social force for simulating moving individuals and SIR model were adopted by our method to simulate the emotional contagion of the crowd. Experimental results showed that the SIR model can effectively improve the fidelity of emotional interaction process and crowd aggregation.
{"title":"Dynamic Crowd Aggregation Simulation Using SIR Model Based Emotion Contagion","authors":"Xiang Nan, Zhou Zehong, Pan Zhigeng","doi":"10.1109/ICVRV.2017.00080","DOIUrl":"https://doi.org/10.1109/ICVRV.2017.00080","url":null,"abstract":"Generating emotion contagion results were very important in crowd simulation field. However, as the individuals often moving dynamically, then computing the contagion process becomes a challenge. In this paper, we focused on the emotion contagion effects on dynamic aggregation process of the virtual pedestrian. Firstly, according to the social force theory, we constructed individuals' moving velocities based on their expectations. Secondly, made an adjacent test to generate the nearer neighbors as emotional contagions usually occurred between neighbors. And then we treated the emotional contagions between individuals and their neighbors as the information spreading process so that we calculated the influences on their moving directions through the emotional information spreading model SIR (Susceptible Infective Removal). Social force for simulating moving individuals and SIR model were adopted by our method to simulate the emotional contagion of the crowd. Experimental results showed that the SIR model can effectively improve the fidelity of emotional interaction process and crowd aggregation.","PeriodicalId":187934,"journal":{"name":"2017 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129336066","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 : 2017-10-01DOI: 10.1109/ICVRV.2017.00048
Yanfen Gan, Jim-Lee Chung, Janson Young
An algorithm that fused block-based algorithm and keypoints-based algorithm is proposed for the copy-move forgery detection. Firstly, the popular keypoint-based algorithms, such as SURF, the nearest-neighbor (2NN) test, adaptive Euclidean distance and Random sample consensus (RANSAC) are applied to extract, match and filter out most of mismatched feature points and get the candidate inlier matches. The RANSAC are addressed to classify the candidate inlier matches. Then, Radial Harmonic Fourier Moments is proposed to extract invariances of the candidate inlier matches in circle blocks. Finally, the host image segment into texture patches. A series of the experiments showed that the proposed fusion algorithm can achieve superior performances than the moment invariant algorithms under various geometric transformations.
{"title":"A Novel Fusion Algorithm for Copy-Move Forgery Detection","authors":"Yanfen Gan, Jim-Lee Chung, Janson Young","doi":"10.1109/ICVRV.2017.00048","DOIUrl":"https://doi.org/10.1109/ICVRV.2017.00048","url":null,"abstract":"An algorithm that fused block-based algorithm and keypoints-based algorithm is proposed for the copy-move forgery detection. Firstly, the popular keypoint-based algorithms, such as SURF, the nearest-neighbor (2NN) test, adaptive Euclidean distance and Random sample consensus (RANSAC) are applied to extract, match and filter out most of mismatched feature points and get the candidate inlier matches. The RANSAC are addressed to classify the candidate inlier matches. Then, Radial Harmonic Fourier Moments is proposed to extract invariances of the candidate inlier matches in circle blocks. Finally, the host image segment into texture patches. A series of the experiments showed that the proposed fusion algorithm can achieve superior performances than the moment invariant algorithms under various geometric transformations.","PeriodicalId":187934,"journal":{"name":"2017 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"56 5-6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132287161","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 : 2017-10-01DOI: 10.1109/ICVRV.2017.00122
Yuanyuan Li, Xun Luo, Edwin Lobo, Andrea Pilco, Yi Chen
This paper proposes wireless ad hoc network simulation based on virtual reality technology. Functional components provided by a game engine such as scene management, animation, rendering, scripting, and physics engine can provide suitable interfaces for ad hoc network simulation. Using these components, simulation of ad hoc networks can be easily implemented. The ad-hoc network simulation technology can be carried out in virtual reality by game engine, which is illustrated by advantages of the game engine and the simulation process in this paper. The workability and accuracy of the simulation scheme are demonstrated, with the metrics of comparing the experimental data with other simulation techniques commonly used.
{"title":"Wireless Ad Hoc Network Simulation Based on Virtual Reality Technology","authors":"Yuanyuan Li, Xun Luo, Edwin Lobo, Andrea Pilco, Yi Chen","doi":"10.1109/ICVRV.2017.00122","DOIUrl":"https://doi.org/10.1109/ICVRV.2017.00122","url":null,"abstract":"This paper proposes wireless ad hoc network simulation based on virtual reality technology. Functional components provided by a game engine such as scene management, animation, rendering, scripting, and physics engine can provide suitable interfaces for ad hoc network simulation. Using these components, simulation of ad hoc networks can be easily implemented. The ad-hoc network simulation technology can be carried out in virtual reality by game engine, which is illustrated by advantages of the game engine and the simulation process in this paper. The workability and accuracy of the simulation scheme are demonstrated, with the metrics of comparing the experimental data with other simulation techniques commonly used.","PeriodicalId":187934,"journal":{"name":"2017 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132356664","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 : 2017-10-01DOI: 10.1109/ICVRV.2017.00022
Xuejin Chen, Haoming Jiang, Tingting Xuan, Lihan Huang, Ligang Liu
Because of its property of saving space, scissor structure, which can transform from a compact state to an expanded state, is widely applied in various fields. In this paper, we solve a challenging problem: designing scissor structures that can expand from a 3D contour to another 3D contour. Given two different 3D shapes specified by users, a non-uniform concentration is required, which makes the problem non-trivial. We propose a three-step algorithm to construct a 3D scissor structure. Firstly we generate scissor segments that are composed of a sequence of planar scissor units based on the shape correspondence. Secondly, we compute the scissor unit parameters of each segment in a suggestive manner. Finally, we introduce ball-shaped joints with parameterized guide slit to realize the deployment in 3D space. The results demonstrate that our algorithm is able to generate scissor structures for a wide range of 3D contours.
{"title":"Scissor-Based 3D Deployable Contours","authors":"Xuejin Chen, Haoming Jiang, Tingting Xuan, Lihan Huang, Ligang Liu","doi":"10.1109/ICVRV.2017.00022","DOIUrl":"https://doi.org/10.1109/ICVRV.2017.00022","url":null,"abstract":"Because of its property of saving space, scissor structure, which can transform from a compact state to an expanded state, is widely applied in various fields. In this paper, we solve a challenging problem: designing scissor structures that can expand from a 3D contour to another 3D contour. Given two different 3D shapes specified by users, a non-uniform concentration is required, which makes the problem non-trivial. We propose a three-step algorithm to construct a 3D scissor structure. Firstly we generate scissor segments that are composed of a sequence of planar scissor units based on the shape correspondence. Secondly, we compute the scissor unit parameters of each segment in a suggestive manner. Finally, we introduce ball-shaped joints with parameterized guide slit to realize the deployment in 3D space. The results demonstrate that our algorithm is able to generate scissor structures for a wide range of 3D contours.","PeriodicalId":187934,"journal":{"name":"2017 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125113122","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 : 2017-10-01DOI: 10.1109/ICVRV.2017.00113
Liqiang Pei, Jinyuan Shen, Runjie Liu
Reducing the dimensionality of datasets is considered an important topic addressed in classification problems. In order to reduce the dimension of the features, a new cascaded method is proposed. Firstly, an improved clustering thought (ICT) is used to screen features initially. Secondly an improved particle swarm optimization (IPSO) in which mutation is adopted into the PSO iteration rule is used to filter out the subsets of features whose value of fitness is larger than the certain threshold. Then the support of each feature can be calculated by these selected sunsets. At last, the best feature subset can be screened according to the sorted support. In order to verify the feasibility of this method, 1588 tobacco leaf images belonging to 41 grades have been experimented. And the experiment results show that the proposed deep feature screening method can effectively improve the image recognition rate and recognition speed.
{"title":"Deep Feature Screening Method by ICT Cascaded with IPSO for Image Recognition","authors":"Liqiang Pei, Jinyuan Shen, Runjie Liu","doi":"10.1109/ICVRV.2017.00113","DOIUrl":"https://doi.org/10.1109/ICVRV.2017.00113","url":null,"abstract":"Reducing the dimensionality of datasets is considered an important topic addressed in classification problems. In order to reduce the dimension of the features, a new cascaded method is proposed. Firstly, an improved clustering thought (ICT) is used to screen features initially. Secondly an improved particle swarm optimization (IPSO) in which mutation is adopted into the PSO iteration rule is used to filter out the subsets of features whose value of fitness is larger than the certain threshold. Then the support of each feature can be calculated by these selected sunsets. At last, the best feature subset can be screened according to the sorted support. In order to verify the feasibility of this method, 1588 tobacco leaf images belonging to 41 grades have been experimented. And the experiment results show that the proposed deep feature screening method can effectively improve the image recognition rate and recognition speed.","PeriodicalId":187934,"journal":{"name":"2017 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129250278","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 : 2017-10-01DOI: 10.1109/ICVRV.2017.00038
Zhike Yi, A. Hao, Wenfeng Song, Hongyi Li, Bowen Li
At present, thyroid cancer has become a serious global public health problem, and ultrasound is the most important imaging method to assess thyroid nodules. But ultrasound diagnostic results of thyroid disease are susceptible to doctors' experiences, levels, status and other factors. So it needs intelligent diagnostic system to assist the doctors to make more objective qualitative and quantitative analyses, to reduce the impact of subjective experience on the diagnostic results. In this paper, a deep learning algorithm for thyroid nodule risk assessment based on ultrasound images is proposed, and an intelligent diagnostic system of thyroid ultrasound image based on this algorithm is constructed. As an aided diagnostic tool, the system is easy to use and can significantly improve the accuracy for determination of thyroid cancer. To verify the effectiveness of the system, we collaborate with the Peking Union Medical College Hospital to test this system.
{"title":"A Novel Intelligent Thyroid Nodule Diagnosis System over Ultrasound Images Based on Deep Learning","authors":"Zhike Yi, A. Hao, Wenfeng Song, Hongyi Li, Bowen Li","doi":"10.1109/ICVRV.2017.00038","DOIUrl":"https://doi.org/10.1109/ICVRV.2017.00038","url":null,"abstract":"At present, thyroid cancer has become a serious global public health problem, and ultrasound is the most important imaging method to assess thyroid nodules. But ultrasound diagnostic results of thyroid disease are susceptible to doctors' experiences, levels, status and other factors. So it needs intelligent diagnostic system to assist the doctors to make more objective qualitative and quantitative analyses, to reduce the impact of subjective experience on the diagnostic results. In this paper, a deep learning algorithm for thyroid nodule risk assessment based on ultrasound images is proposed, and an intelligent diagnostic system of thyroid ultrasound image based on this algorithm is constructed. As an aided diagnostic tool, the system is easy to use and can significantly improve the accuracy for determination of thyroid cancer. To verify the effectiveness of the system, we collaborate with the Peking Union Medical College Hospital to test this system.","PeriodicalId":187934,"journal":{"name":"2017 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114069595","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 : 2017-10-01DOI: 10.1109/ICVRV.2017.00098
Hongqiang Zhu, Jing Tong, Luosheng Zhang, X. Zou
A mobile-based virtual try-on system is proposed to deal with the problems of high cost and conflicts between computational complexity and simulation effects. In this paper, several modules are included, such as automatic 3D face reconstruction based on a single image, auto-skinning and realtime local simulation of cloth. According to the experiments, the virtual try-on system introduced in this paper is able to achieve better fitting effects with lower constructing and computing costs, in which case good experience of mobile-based virtual try-on system is provided.
{"title":"Research and Development of Virtual Try-On System Based on Mobile Platform","authors":"Hongqiang Zhu, Jing Tong, Luosheng Zhang, X. Zou","doi":"10.1109/ICVRV.2017.00098","DOIUrl":"https://doi.org/10.1109/ICVRV.2017.00098","url":null,"abstract":"A mobile-based virtual try-on system is proposed to deal with the problems of high cost and conflicts between computational complexity and simulation effects. In this paper, several modules are included, such as automatic 3D face reconstruction based on a single image, auto-skinning and realtime local simulation of cloth. According to the experiments, the virtual try-on system introduced in this paper is able to achieve better fitting effects with lower constructing and computing costs, in which case good experience of mobile-based virtual try-on system is provided.","PeriodicalId":187934,"journal":{"name":"2017 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123792885","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：