In this paper we introduce a new reverse Loop subdivision method. In contrast to current wavelets based Loop subdivision scheme, our method applies the same rules to both regular and extraordinary vertices and reconst...In this paper we introduce a new reverse Loop subdivision method. In contrast to current wavelets based Loop subdivision scheme, our method applies the same rules to both regular and extraordinary vertices and reconstructs the sharp features easily. Furthermore, our method runs faster because it does not need analysis and synthesis procedural. Our main goal is the design of a reverse subdivision method that can reconstruct the coarser mesh from a finer subdivision surface with sharp features for multiresolution representation, The proposed method only needs a little memory storage and brings little error, and it is easy to implement.展开更多
Due to the shortages of current methods for the recovery of sharp features of mesh models with holes,this paper presents two novel algorithms for the recovery of features(especially sharp features)in mesh models.One a...Due to the shortages of current methods for the recovery of sharp features of mesh models with holes,this paper presents two novel algorithms for the recovery of features(especially sharp features)in mesh models.One algorithm defines an energy that is regarded as the difference between the initial features and the ideal features.The optimal solution of the energy optimization problem modifies the initial features.The algorithm has good performance on sharp features.The other method establishes a plane cluster for each initial feature point to obtain a corresponding modified feature point.If necessary,we can obtain the modified feature line by fitting these modified points.Both methods depend little on the result of fillingmodel holes and result in better features,which maintain the sharp geometric characteristic and the smoothness of the model.The experimental results of the two algorithms demonstrate their superiority and rationality compared with the existing methods.展开更多
We present a robust mesh sharpening approach to reconstructing sharp features from blended or chamfered features, even with noise and aliasing errors. Feature regions were first recognized via normal variation accordi...We present a robust mesh sharpening approach to reconstructing sharp features from blended or chamfered features, even with noise and aliasing errors. Feature regions were first recognized via normal variation according to the user's input, and then normal filtering was applied to faces of feature regions. Finally, the vertices of the feature region were gradually updated based on new face normals using a least-squares error criterion. Experimental results demonstrate that the method is effective and robust in sharpening meshes.展开更多
This paper presents a reconstruction algorithm to build a surface mesh approximating an object from an unorganized point sampling of the boundary object. It combines 3D Delaunay tetrahedralization and mesh-growing met...This paper presents a reconstruction algorithm to build a surface mesh approximating an object from an unorganized point sampling of the boundary object. It combines 3D Delaunay tetrahedralization and mesh-growing method and uses only once Delau- nay triangulation. It begins with 3D Delaunay triangulation of the sampling. Then initialize the surface mesh with seed facets se- lected from Delaunay triangulation. Selection is based on the angle formed by the circumscribing ball of incident tetrahedral. Finally, grow until complete the surface mesh based on some heuristic rules. This paper shows several experimental results that demonstrate this method can handle open and close surfaces and work efficiently on various object topologies except non-manifold surface with self-intersections. It can reproduce even the smallest details of well-sampled surfaces but not work properly in every under-sampled situation that point density is too low.展开更多
Sensor data, typically images and laser data, are essential to modeling real plants. However, due to the complex geometry of the plants, the measurement data are generally limited, thereby bringing great difficulties ...Sensor data, typically images and laser data, are essential to modeling real plants. However, due to the complex geometry of the plants, the measurement data are generally limited, thereby bringing great difficulties in classifying and constructing plant organs, comprising leaves and branches. The paper presents an approach to modeling plants with the sensor data by detecting reliable sharp features, i.e. the leaf apexes of the plants with leaves and the branch tips of the plants without leaves, on volumes recovered from the raw data. The extracted features provide good estimations of correct positions of the organs. Thereafter, the leaves are reconstructed separately by simply fitting and optimizing a generic leaf model. One advantage of the method is that it involves limited manual intervention. For plants without leaves, we develop an efficient strategy for decomposition-based skeletonization by using the tip features to reconstruct the 3D models from noisy laser data. Experiments show that the sharp feature detection algorithm is effective, and the proposed plant modeling approach is competent in constructing realistic models with sensor data.展开更多
基金Supported by the High Technology Research and Development Progrmnn~ of China （No. 2003AA411310）, the National Natural Science Foundation of China （No. 60373070） and Microsoft Research Project 2005-1.
文摘In this paper we introduce a new reverse Loop subdivision method. In contrast to current wavelets based Loop subdivision scheme, our method applies the same rules to both regular and extraordinary vertices and reconstructs the sharp features easily. Furthermore, our method runs faster because it does not need analysis and synthesis procedural. Our main goal is the design of a reverse subdivision method that can reconstruct the coarser mesh from a finer subdivision surface with sharp features for multiresolution representation, The proposed method only needs a little memory storage and brings little error, and it is easy to implement.
基金The authors are supported by a NKBRPC(2011CB302400)the National Natural Science Foundation of China(11171322 and 11371341)the 111 Project(No.b07033).
文摘Due to the shortages of current methods for the recovery of sharp features of mesh models with holes,this paper presents two novel algorithms for the recovery of features(especially sharp features)in mesh models.One algorithm defines an energy that is regarded as the difference between the initial features and the ideal features.The optimal solution of the energy optimization problem modifies the initial features.The algorithm has good performance on sharp features.The other method establishes a plane cluster for each initial feature point to obtain a corresponding modified feature point.If necessary,we can obtain the modified feature line by fitting these modified points.Both methods depend little on the result of fillingmodel holes and result in better features,which maintain the sharp geometric characteristic and the smoothness of the model.The experimental results of the two algorithms demonstrate their superiority and rationality compared with the existing methods.
基金supported by the Hi-Tech Research and Development Pro-gram (863) of China (Nos. 2007AA01Z311 and 2007AA04Z1A5)the Doctoral Fund of MOE of China (No. 20060335114)the Science and Technology Program of Zhejiang Province, China (No. 2007C21006)
文摘We present a robust mesh sharpening approach to reconstructing sharp features from blended or chamfered features, even with noise and aliasing errors. Feature regions were first recognized via normal variation according to the user's input, and then normal filtering was applied to faces of feature regions. Finally, the vertices of the feature region were gradually updated based on new face normals using a least-squares error criterion. Experimental results demonstrate that the method is effective and robust in sharpening meshes.
基金Supported by National Natural Science Foundation of China(No.60875046)Program for Changjiang Scholars and Innovative Research Team in University(No.IRT1109)+5 种基金the Key Project of Chinese Ministry of Education(No.209029)the Program for Liaoning Excellent Talents in University(No.LR201003)the Program for Liaoning Science and Technology Research in University(No.LS2010008,2009S008,2009S009, LS2010179)the Program for Liaoning Innovative Research Team in University(Nos.2009T005, LT2010005, LT2011018)Natural Science Foundation of Liaoning Province(201102008)"Liaoning Bai Qian Wan Talents Program(2010921010, 2011921009)"
文摘This paper presents a reconstruction algorithm to build a surface mesh approximating an object from an unorganized point sampling of the boundary object. It combines 3D Delaunay tetrahedralization and mesh-growing method and uses only once Delau- nay triangulation. It begins with 3D Delaunay triangulation of the sampling. Then initialize the surface mesh with seed facets se- lected from Delaunay triangulation. Selection is based on the angle formed by the circumscribing ball of incident tetrahedral. Finally, grow until complete the surface mesh based on some heuristic rules. This paper shows several experimental results that demonstrate this method can handle open and close surfaces and work efficiently on various object topologies except non-manifold surface with self-intersections. It can reproduce even the smallest details of well-sampled surfaces but not work properly in every under-sampled situation that point density is too low.
基金Supported in part by the National Basic Research Program of China (Grant No. 2004CB318000)the National High-Tech Research & Development Program of China (Grant Nos. 2006AA01Z301, 2006AA01Z302, 2007AA01Z336)Key Grant Project of Chinese Ministry of Education (Grant No. 103001)
文摘Sensor data, typically images and laser data, are essential to modeling real plants. However, due to the complex geometry of the plants, the measurement data are generally limited, thereby bringing great difficulties in classifying and constructing plant organs, comprising leaves and branches. The paper presents an approach to modeling plants with the sensor data by detecting reliable sharp features, i.e. the leaf apexes of the plants with leaves and the branch tips of the plants without leaves, on volumes recovered from the raw data. The extracted features provide good estimations of correct positions of the organs. Thereafter, the leaves are reconstructed separately by simply fitting and optimizing a generic leaf model. One advantage of the method is that it involves limited manual intervention. For plants without leaves, we develop an efficient strategy for decomposition-based skeletonization by using the tip features to reconstruct the 3D models from noisy laser data. Experiments show that the sharp feature detection algorithm is effective, and the proposed plant modeling approach is competent in constructing realistic models with sensor data.
