Y. Pan, Mingquan Zhou, Yachun Fan, Dongdong Zhang, Xia Zheng
Image-based modeling problem is a tough and basic problem in computer vision and computer graphics. This paper provides an improved solution to generate a scene model from a single image based on the multi-scale Markov random field. We consider the different influence of the different color components in RGB color space when modeling through a single still image. In this paper, we proposed a weighted color model for 3D reconstruction and draw a conclusion that different color components contain different depth information, and by modifying the weight of different color components in the image we can get a better visual effects and more accurate 3D model. The output model of our solution then can be used to construct a virtual scene rapidly based on a scene image.
{"title":"A Weighted Color MRF Model for 3D Reconstruction from a Single Image","authors":"Y. Pan, Mingquan Zhou, Yachun Fan, Dongdong Zhang, Xia Zheng","doi":"10.1109/ICVRV.2013.12","DOIUrl":"https://doi.org/10.1109/ICVRV.2013.12","url":null,"abstract":"Image-based modeling problem is a tough and basic problem in computer vision and computer graphics. This paper provides an improved solution to generate a scene model from a single image based on the multi-scale Markov random field. We consider the different influence of the different color components in RGB color space when modeling through a single still image. In this paper, we proposed a weighted color model for 3D reconstruction and draw a conclusion that different color components contain different depth information, and by modifying the weight of different color components in the image we can get a better visual effects and more accurate 3D model. The output model of our solution then can be used to construct a virtual scene rapidly based on a scene image.","PeriodicalId":179465,"journal":{"name":"2013 International Conference on Virtual Reality and Visualization","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132850961","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}
Jiamin Liu, Shuo Chen, Hongxing Sun, Y. Qin, Xibo Wang
To meet the user demands and to enhance integrating of markers into the environment that is close to nature as much as possible, this paper proposes a color marker-based method for tracking the target object in real time and registering a virtual three-dimensional (3D) object. Firstly, the contours of the color marker patterns on the video images are extracted by using adaptive threshold. Each marker pattern on the image is corrected and recognized by the template pattern. Secondly, the specified color markers are identified by using the L1L2L3 and rgb models. Finally, the virtual objects are registered in real-time on the marker image by using the camera parameters derived from data of the color markers. The experimental results show that the virtual objects can be accurately registered on the markers of the video images. Compared to the Ar Toolkit software, this method is able to work in the environment with various illumination conditions.
{"title":"Real Time Tracking Method by Using Color Markers","authors":"Jiamin Liu, Shuo Chen, Hongxing Sun, Y. Qin, Xibo Wang","doi":"10.1109/ICVRV.2013.25","DOIUrl":"https://doi.org/10.1109/ICVRV.2013.25","url":null,"abstract":"To meet the user demands and to enhance integrating of markers into the environment that is close to nature as much as possible, this paper proposes a color marker-based method for tracking the target object in real time and registering a virtual three-dimensional (3D) object. Firstly, the contours of the color marker patterns on the video images are extracted by using adaptive threshold. Each marker pattern on the image is corrected and recognized by the template pattern. Secondly, the specified color markers are identified by using the L1L2L3 and rgb models. Finally, the virtual objects are registered in real-time on the marker image by using the camera parameters derived from data of the color markers. The experimental results show that the virtual objects can be accurately registered on the markers of the video images. Compared to the Ar Toolkit software, this method is able to work in the environment with various illumination conditions.","PeriodicalId":179465,"journal":{"name":"2013 International Conference on Virtual Reality and Visualization","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114539476","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}
A parallel adaptive ray-casting volume rendering algorithm based on information entropy is presented in this paper, where information entropy is introduced to measure the distribution of physical features in a data field and accordingly guide an effective data sampling. The algorithm adopts a patch-wise adaptation: Different sampling rates are assigned to the data patches according to their information entropies while uniform sampling is employed within each data patch. A composing method is designed subsequently for non-uniform sampling points on each ray. The experiments show that compared with the uniform rendering algorithm, the presented adaptive rendering algorithm can achieve a rendering speedup ratio of 1.4~2.1 in a distributed-memory environment.
{"title":"A Parallel Adaptive Volume Rendering Algorithm Based on Information Entropy","authors":"Huawei Wang, Yi Cao, Li Xiao, Guoqing Wu","doi":"10.1109/ICVRV.2013.28","DOIUrl":"https://doi.org/10.1109/ICVRV.2013.28","url":null,"abstract":"A parallel adaptive ray-casting volume rendering algorithm based on information entropy is presented in this paper, where information entropy is introduced to measure the distribution of physical features in a data field and accordingly guide an effective data sampling. The algorithm adopts a patch-wise adaptation: Different sampling rates are assigned to the data patches according to their information entropies while uniform sampling is employed within each data patch. A composing method is designed subsequently for non-uniform sampling points on each ray. The experiments show that compared with the uniform rendering algorithm, the presented adaptive rendering algorithm can achieve a rendering speedup ratio of 1.4~2.1 in a distributed-memory environment.","PeriodicalId":179465,"journal":{"name":"2013 International Conference on Virtual Reality and Visualization","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117236803","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}
In this paper we propose a rotation invariant face detection algorithm to detect the face on the mobile devices. Due to the performance limit on the mobile devices, complex face detection process must be avoided in order to generate a fast detection. Most proposed face detection algorithms are highly rotational variant, which means only upright face can be detected in the image, and those existed rotational invariant face detection algorithms are too complicated to be suitable for the mobile devices. Our algorithm can detect non-upright face in the image with an efficient process implemented on the mobile device (Google Nexus 7 tablet), and has different detection ranges adapting to different user choices. The experiment results reveal the validity of our proposed algorithm. Meanwhile, our algorithm can be easily implemented on other mobile systems, such as Windows Phone, iPhone and so on.
{"title":"Rotational Invariant Face Detection on a Mobile Device","authors":"Guanghui Ma, Weiliang Meng, Shibiao Xu, Xiaopeng Zhang","doi":"10.1109/ICVRV.2013.66","DOIUrl":"https://doi.org/10.1109/ICVRV.2013.66","url":null,"abstract":"In this paper we propose a rotation invariant face detection algorithm to detect the face on the mobile devices. Due to the performance limit on the mobile devices, complex face detection process must be avoided in order to generate a fast detection. Most proposed face detection algorithms are highly rotational variant, which means only upright face can be detected in the image, and those existed rotational invariant face detection algorithms are too complicated to be suitable for the mobile devices. Our algorithm can detect non-upright face in the image with an efficient process implemented on the mobile device (Google Nexus 7 tablet), and has different detection ranges adapting to different user choices. The experiment results reveal the validity of our proposed algorithm. Meanwhile, our algorithm can be easily implemented on other mobile systems, such as Windows Phone, iPhone and so on.","PeriodicalId":179465,"journal":{"name":"2013 International Conference on Virtual Reality and Visualization","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123946674","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}
Craniofacial superimposition is a forensic process that aims to identify a missing person from a photograph and an unknown dead's skull. One of the crucial steps is skull-face overlay in consistent with photographic space according to the face photo. Thus parameter estimation of perspective projection becomes key problem. In this paper, we employ a method of camera calibration based on vanishing point(VPBC) to estimate relevant parameters including perspective angle, orientation of camera and photographic object distance. Firstly, according to two vanishing points in image calculate the perspective angle and focal length using photo-geography theory, then evaluate camera direction and photographic object distance based on equal focal length, finally, map the 3D skull according to above known parameters, and implement skull-face overlay process. Result shows that this approach is effective, efficient and accurate for follow-up research work.
{"title":"Parameter Estimation for Perspective Projection Based on Camera Calibration in Skull-Face Overlay","authors":"Wuxia Jin, Guohua Geng, Kang Li, Yi Han","doi":"10.1109/ICVRV.2013.62","DOIUrl":"https://doi.org/10.1109/ICVRV.2013.62","url":null,"abstract":"Craniofacial superimposition is a forensic process that aims to identify a missing person from a photograph and an unknown dead's skull. One of the crucial steps is skull-face overlay in consistent with photographic space according to the face photo. Thus parameter estimation of perspective projection becomes key problem. In this paper, we employ a method of camera calibration based on vanishing point(VPBC) to estimate relevant parameters including perspective angle, orientation of camera and photographic object distance. Firstly, according to two vanishing points in image calculate the perspective angle and focal length using photo-geography theory, then evaluate camera direction and photographic object distance based on equal focal length, finally, map the 3D skull according to above known parameters, and implement skull-face overlay process. Result shows that this approach is effective, efficient and accurate for follow-up research work.","PeriodicalId":179465,"journal":{"name":"2013 International Conference on Virtual Reality and Visualization","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130301584","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}
Ma Zhiqiang, Wang Lili, Zhang Xinwei, W. Ke, Zhao Qin-ping
In remote visualization for complex 3D dynamic datasets, existing vertex compression algorithms have low compression factor, and compressed datasets cannot be transmitted to client in once if its size is bigger than bandwidth. In order to solve these problems, this paper presents a compact impression for 3D dynamic datasets in certain range. We construct an equilateral triangle to obtain and compress parts of original datasets, which realizes rigid body decomposition in certain range. In order to promote compression factor, we also introduce a temporal rigid body decomposition to group vertices with temporal trajectory consistency. The results of the experiments show that our algorithm can obtain high compression factor for dynamic datasets in certain range to reduce limitation of bandwidth.
{"title":"Compact Compression for 3D Dynamic Datasets in Certain Range","authors":"Ma Zhiqiang, Wang Lili, Zhang Xinwei, W. Ke, Zhao Qin-ping","doi":"10.1109/ICVRV.2013.13","DOIUrl":"https://doi.org/10.1109/ICVRV.2013.13","url":null,"abstract":"In remote visualization for complex 3D dynamic datasets, existing vertex compression algorithms have low compression factor, and compressed datasets cannot be transmitted to client in once if its size is bigger than bandwidth. In order to solve these problems, this paper presents a compact impression for 3D dynamic datasets in certain range. We construct an equilateral triangle to obtain and compress parts of original datasets, which realizes rigid body decomposition in certain range. In order to promote compression factor, we also introduce a temporal rigid body decomposition to group vertices with temporal trajectory consistency. The results of the experiments show that our algorithm can obtain high compression factor for dynamic datasets in certain range to reduce limitation of bandwidth.","PeriodicalId":179465,"journal":{"name":"2013 International Conference on Virtual Reality and Visualization","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129882466","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}
Geographic network visualization lay out the data with respect to the physical location of nodes in the network. This paper presents a method for visualizing network communication in geographic environment. The method first simplifies GIS data and aggregates network communication data according to both geographic distribution and network topology, and then followed by selecting and highlighting main communication links. Finally, the layout of some nodes is adjusted in the network. By using our method, the network communication data can be visualized in geographic environment aesthetically and intelligibly. We developed a prototype system to help analysts and users understanding the characteristics and patterns of the networks communication.
{"title":"Visualizing Network Communication in Geographic Environment","authors":"Haifeng Hu, Hengyuan Zhang, Wei Li","doi":"10.1109/ICVRV.2013.39","DOIUrl":"https://doi.org/10.1109/ICVRV.2013.39","url":null,"abstract":"Geographic network visualization lay out the data with respect to the physical location of nodes in the network. This paper presents a method for visualizing network communication in geographic environment. The method first simplifies GIS data and aggregates network communication data according to both geographic distribution and network topology, and then followed by selecting and highlighting main communication links. Finally, the layout of some nodes is adjusted in the network. By using our method, the network communication data can be visualized in geographic environment aesthetically and intelligibly. We developed a prototype system to help analysts and users understanding the characteristics and patterns of the networks communication.","PeriodicalId":179465,"journal":{"name":"2013 International Conference on Virtual Reality and Visualization","volume":"12 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128372975","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}
Afeng Yang, Jianfei Wu, Min Lu, Shuhua Teng, Jixiang Sun
Blind Restoration of adaptive optics images is important in the field of astronomical imaging and space object surveillance. Using multi frame blind deconvolution as main technique means for high resolution restoration, a general cost function is deduced to deconvolve Poisson noise model image under the Bayesian-MAP estimate framework. To minimize the cost function, a solution algorithm based on alternating recursion method is proposed. In addition, asymmetric iteration method is introduced into solution process to avoid converging to local minima and maintain robustness of restored image. Experimental results show that the proposed method can recover high quality image from turbulence degraded images effectively and alleviate the negative influence of noise on the restoration result.
{"title":"Asymmetric Multiframe Blind Restoration for Adaptive Optics Images via Alternating Recursion","authors":"Afeng Yang, Jianfei Wu, Min Lu, Shuhua Teng, Jixiang Sun","doi":"10.1109/ICVRV.2013.35","DOIUrl":"https://doi.org/10.1109/ICVRV.2013.35","url":null,"abstract":"Blind Restoration of adaptive optics images is important in the field of astronomical imaging and space object surveillance. Using multi frame blind deconvolution as main technique means for high resolution restoration, a general cost function is deduced to deconvolve Poisson noise model image under the Bayesian-MAP estimate framework. To minimize the cost function, a solution algorithm based on alternating recursion method is proposed. In addition, asymmetric iteration method is introduced into solution process to avoid converging to local minima and maintain robustness of restored image. Experimental results show that the proposed method can recover high quality image from turbulence degraded images effectively and alleviate the negative influence of noise on the restoration result.","PeriodicalId":179465,"journal":{"name":"2013 International Conference on Virtual Reality and Visualization","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126873889","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}
Given a low-quality depth video and its corresponding high-quality color video, we intend to improve depth quality by suppressing both spatial and temporal noise. A refined weighted mode filtering method (WMF) based on a joint histogram is proposed. For WMF, similarity between reference and neighbor pixels plays an important role in counting each bin of the joint histogram. Since calculating similarity using single pixel will be affected by random pixel noise, we utilize patch-based NL-means (Non-Local means) for structure-aware similarity calculation, also, we fuse color and depth similarity adaptively with credibility maps to deal with texture copying problem. For temporally consistent recovery, we introduce inter frame correlation by integrating neighboring frames with optical flow and patch-based similarity measurement. Experimental results show that our proposed method has achieved more complete and clear depth, especially in discontinuous areas. Furthermore, temporally enhancement of depth video addresses flickering problem and gets more stable depth.
{"title":"A Refined Weighted Mode Filtering Approach for Depth Video Enhancement","authors":"X. Zuo, Jiangbin Zheng","doi":"10.1109/ICVRV.2013.30","DOIUrl":"https://doi.org/10.1109/ICVRV.2013.30","url":null,"abstract":"Given a low-quality depth video and its corresponding high-quality color video, we intend to improve depth quality by suppressing both spatial and temporal noise. A refined weighted mode filtering method (WMF) based on a joint histogram is proposed. For WMF, similarity between reference and neighbor pixels plays an important role in counting each bin of the joint histogram. Since calculating similarity using single pixel will be affected by random pixel noise, we utilize patch-based NL-means (Non-Local means) for structure-aware similarity calculation, also, we fuse color and depth similarity adaptively with credibility maps to deal with texture copying problem. For temporally consistent recovery, we introduce inter frame correlation by integrating neighboring frames with optical flow and patch-based similarity measurement. Experimental results show that our proposed method has achieved more complete and clear depth, especially in discontinuous areas. Furthermore, temporally enhancement of depth video addresses flickering problem and gets more stable depth.","PeriodicalId":179465,"journal":{"name":"2013 International Conference on Virtual Reality and Visualization","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121881905","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}
We propose a new efficient technique for acquiring normal maps of nearly flat objects using a flatbed scanner. The technique requires no hardware modification or dedicated calibration. Our setup consists of a conventional flatbed scanner and a clamp on which a planar checkerboard is glued. We take four scans of the object manually rotated by roughly 90 degrees. The scans are then registered by sub pixel checkerboard corners. Detailed normal map of the scanned surface is obtained by solving a linear least squares problem.
{"title":"Normal Map Acquisition of Nearly Flat Objects Using a Flatbed Scanner","authors":"Rongjiang Pan, V. Skala","doi":"10.1109/ICVRV.2013.19","DOIUrl":"https://doi.org/10.1109/ICVRV.2013.19","url":null,"abstract":"We propose a new efficient technique for acquiring normal maps of nearly flat objects using a flatbed scanner. The technique requires no hardware modification or dedicated calibration. Our setup consists of a conventional flatbed scanner and a clamp on which a planar checkerboard is glued. We take four scans of the object manually rotated by roughly 90 degrees. The scans are then registered by sub pixel checkerboard corners. Detailed normal map of the scanned surface is obtained by solving a linear least squares problem.","PeriodicalId":179465,"journal":{"name":"2013 International Conference on Virtual Reality and Visualization","volume":"163 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124569947","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：