基于多特征融合的三维形状分割方法
|
摘要
三维形状分割是三维形状分析中的一个重要问题.针对单一特征对同一类模型分割结果存在较大差异的问题,提出一种基于学习的多特征融合的三维形状分割方法.首先利用过分割方法将三维模型分割成多个子面片,分别对每个子面片提取多种几何特征;然后将几何特征作为低层特征输入深度神经网络模型,通过学习生成高层特征;最后基于该高层特征用高斯混合模型的方法得到聚类中心,利用图割得到最后分割结果.在普林斯顿标准数据集和COSEG数据集上的实验结果表明,与传统分割方法相比,该方法具有较好的一致性分割结果.
3D shape segmentation is an important issue in shape analysis. This paper proposed a segmentation method based on multi-feature fusion to solve the consistency problem. A 3D model is first divided into multiple sub-patches by using over-segmentation. Then we use the geometric features extracted from each sub-patch as low-level features input for depth neural network model to generate high-level features. Finally, based on these high-level features, Gaussian mixture model is employed to get the clustering centers and graph-cut is adapted for the final segmentation. Experiments on PSB and COSEG datasets show that the proposed method outperforms the traditional geometric feature method, and can get good consistency results for the same kind of 3D shapes.
3D shape segmentation is an important issue in shape analysis. This paper proposed a segmentation method based on multi-feature fusion to solve the consistency problem. A 3D model is first divided into multiple sub-patches by using over-segmentation. Then we use the geometric features extracted from each sub-patch as low-level features input for depth neural network model to generate high-level features. Finally, based on these high-level features, Gaussian mixture model is employed to get the clustering centers and graph-cut is adapted for the final segmentation. Experiments on PSB and COSEG datasets show that the proposed method outperforms the traditional geometric feature method, and can get good consistency results for the same kind of 3D shapes.
引文
[1]Wu Z Z,Wang Y H,Shou R Y,et al.Unsupervised co-segmentation of 3D shapes via affinity aggregation spectral clustering[J].Computers&Graphics,2013,37(6):628-637
[2]Bronstein M M,Kokkinos I.Scale-invariant heat kernel signatures for non-rigid shape recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2010:1704-1711
[3]Shapira L,Shamir A,Cohen-Or D.Consistent mesh partitioning and skeletonisation using the shape diameter function[J].The Visual Computer,2008,24(4):249
[4]Gal R,Cohen-Or D.Salient geometric features for partial shape matching and similarity[J].ACM Transactions on Graphics,2006,25(1):130-150
[5]Katz S,Tal A.Hierarchical mesh decomposition using fuzzy clustering and cuts[M].New York:ACM Press,2003
[6]Li C Y,Hamza A B.A multiresolution descriptor for deformable 3D shape retrieval[J].The Visual Computer,2013,29(6-8):513-524
[7]Wang Y H,Asafi S,van Kaick O,et al.Active co-analysis of a set of shapes[J].ACM Transactions on Graphics,2012,31(6):Article No.165
[8]Ben-Chen M,Gotsman C.Characterizing shape using conformal factors[C]//Proceedings of the 1st Eurographics Conference on3D Object Retrieval.Aire-la-Ville:Eurographics Association Press,2008:1-8
[9]Meng M,Xia J Z,Luo J,et al.Unsupervised co-segmentation for3D shapes using iterative multi-label optimization[J].Computer-Aided Design,2013,45(2):312-320
[10]Golovinskiy A,Funkhouser T.Randomized cuts for 3D mesh analysis[J].ACM Transactions on Graphics,2008,27(5):Article No.145
[11]Shamir A.A survey on mesh segmentation tecniques[J].Computer Graphics Forum,2008,27(6):1539-1556
[12]van Kaick O,Fish N,Kleiman Y,et al.Shape segmentation by approximate convexity analysis[J].ACM Transactions on Graphics,2014,34(1):Article No.4
[13]Zhou Y,Yin K X,Huang H,et al.Generalized cylinder decomposition[J].ACM Transactions on Graphics,2015,34(6):Article No.171
[14]Xie Z G,Xu K,Liu L G,et al.3D shape segmentation and labeling via extreme learning machine[J].Computer Graphics Forum,2014,33(5):85-95
[15]Liu R,Zhang H.Segmentation of 3D meshes through spectral clustering[C]//Proceedings of the 12th Pacific Conference on Computer Graphics and Applications.Los Alamitos:IEEEComputer Society Press,2004:298-305
[16]Shlafman S,Tal A,Katz S.Metamorphosis of polyhedral surfaces using decomposition[J].Computer Graphics Forum,2002,21(3):219-228
[17]Lai Y K,Hu S M,Martin R R,et al.Fast mesh segmentation using random walks[C]//Proceedings of the ACM Symposium on Solid and Physical Modeling.New York:ACM Press,2008:183-191
[18]Chen X B,Golovinskiy A,Funkhouser T.A benchmark for 3Dmesh segmentation[J].ACM Transactions on Graphics,2009,28(3):Article No.73
[19]Golovinskiy A,Funkhouser T.Consistent segmentation of 3Dmodels[J].Computers&Graphics,2009,33(3):262-269
[20]Xu K,Li H H,Zhang H,et al.Style-content separation by anisotropic part scales[J].ACM Transactions on Graphics,2010,29(6):Article No.184
[21]Zheng Y Y,Cohen Or D,Averkiou M,et al.Recurring part arrangements in shape collections[J].Computer Graphics Forum,2014,33(2):115-124
[22]Kalogerakis E,Hertzmann A,Singh K.Learning 3D mesh segmentation and labeling[J].ACM Transactions on Graphics,2010,29(4):Article No.102
[23]Sidi O,van Kaick O,Kleiman Y,et al.Unsupervised co-segmentation of a set of shapes via descriptor-space spectral clustering[J].ACM Transactions on Graphics,2011,30(6):Article No.126
[24]Huang Q X,Koltun V,Guibas L.Joint shape segmentation with linear programming[J].ACM Transactions on Graphics,2011,30(6):Article No.125
[25]Hu R Z,Fan L B,Liu L G.Co-segmentation of 3D shapes via subspace clustering[J].Computer Graphics Forum,2012,31(5):1703-1713
[26]Gatzke T,Grimm C,Garland M,et al.Curvature maps for local shape comparison[C]//Proceedings of International Conference on Shape Modeling and Applications.Los Alamitos:IEEEComputer Society Press,2005:244-253
[27]Vranic D V.An improvement of rotation invariant 3D-shape based on functions on concentric spheres[C]//Proceedings of International Conference on Image Processing.Los Alamitos:IEEE Computer Society Press,2003,3:757-760
[28]Guo K,Zou D G,Chen X W.3D mesh labeling via deep convolutional neural networks[J].ACM Transactions on Graphics,2015,35(1):Article No.3
[29]Ren X,Malik J.Learning a classification model for segmentation[C]//Proceeding of the 9th IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2003,1:10-17
[30]Shi J,Malik J.Normalized cuts and image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8):888-905
[31]LavouéG.Bag of words and local spectral descriptor for 3Dpartial shape retrieval[C]//Proceedings of the 4th Eurographics Conference on 3D Object Retrieval.Aire-la-Ville:Eurographics Association Press,2011:41-48
[32]Bengio Y,Lamblin P,Popovici D,et al.Greedy layer-wise training of deep networks[C]//Proceedings of Annual Conference on Neural Information Processing Systems.Cambridge:MIT Press,2007:153-160
[33]Pérez-Cruz F.Kullback-Leibler divergence estimation of continuous distributions[C]//Proceedings of IEEE International Symposium on Information Theory.Los Alamitos:IEEEComputer Society Press,2008:1666-1670
[2]Bronstein M M,Kokkinos I.Scale-invariant heat kernel signatures for non-rigid shape recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2010:1704-1711
[3]Shapira L,Shamir A,Cohen-Or D.Consistent mesh partitioning and skeletonisation using the shape diameter function[J].The Visual Computer,2008,24(4):249
[4]Gal R,Cohen-Or D.Salient geometric features for partial shape matching and similarity[J].ACM Transactions on Graphics,2006,25(1):130-150
[5]Katz S,Tal A.Hierarchical mesh decomposition using fuzzy clustering and cuts[M].New York:ACM Press,2003
[6]Li C Y,Hamza A B.A multiresolution descriptor for deformable 3D shape retrieval[J].The Visual Computer,2013,29(6-8):513-524
[7]Wang Y H,Asafi S,van Kaick O,et al.Active co-analysis of a set of shapes[J].ACM Transactions on Graphics,2012,31(6):Article No.165
[8]Ben-Chen M,Gotsman C.Characterizing shape using conformal factors[C]//Proceedings of the 1st Eurographics Conference on3D Object Retrieval.Aire-la-Ville:Eurographics Association Press,2008:1-8
[9]Meng M,Xia J Z,Luo J,et al.Unsupervised co-segmentation for3D shapes using iterative multi-label optimization[J].Computer-Aided Design,2013,45(2):312-320
[10]Golovinskiy A,Funkhouser T.Randomized cuts for 3D mesh analysis[J].ACM Transactions on Graphics,2008,27(5):Article No.145
[11]Shamir A.A survey on mesh segmentation tecniques[J].Computer Graphics Forum,2008,27(6):1539-1556
[12]van Kaick O,Fish N,Kleiman Y,et al.Shape segmentation by approximate convexity analysis[J].ACM Transactions on Graphics,2014,34(1):Article No.4
[13]Zhou Y,Yin K X,Huang H,et al.Generalized cylinder decomposition[J].ACM Transactions on Graphics,2015,34(6):Article No.171
[14]Xie Z G,Xu K,Liu L G,et al.3D shape segmentation and labeling via extreme learning machine[J].Computer Graphics Forum,2014,33(5):85-95
[15]Liu R,Zhang H.Segmentation of 3D meshes through spectral clustering[C]//Proceedings of the 12th Pacific Conference on Computer Graphics and Applications.Los Alamitos:IEEEComputer Society Press,2004:298-305
[16]Shlafman S,Tal A,Katz S.Metamorphosis of polyhedral surfaces using decomposition[J].Computer Graphics Forum,2002,21(3):219-228
[17]Lai Y K,Hu S M,Martin R R,et al.Fast mesh segmentation using random walks[C]//Proceedings of the ACM Symposium on Solid and Physical Modeling.New York:ACM Press,2008:183-191
[18]Chen X B,Golovinskiy A,Funkhouser T.A benchmark for 3Dmesh segmentation[J].ACM Transactions on Graphics,2009,28(3):Article No.73
[19]Golovinskiy A,Funkhouser T.Consistent segmentation of 3Dmodels[J].Computers&Graphics,2009,33(3):262-269
[20]Xu K,Li H H,Zhang H,et al.Style-content separation by anisotropic part scales[J].ACM Transactions on Graphics,2010,29(6):Article No.184
[21]Zheng Y Y,Cohen Or D,Averkiou M,et al.Recurring part arrangements in shape collections[J].Computer Graphics Forum,2014,33(2):115-124
[22]Kalogerakis E,Hertzmann A,Singh K.Learning 3D mesh segmentation and labeling[J].ACM Transactions on Graphics,2010,29(4):Article No.102
[23]Sidi O,van Kaick O,Kleiman Y,et al.Unsupervised co-segmentation of a set of shapes via descriptor-space spectral clustering[J].ACM Transactions on Graphics,2011,30(6):Article No.126
[24]Huang Q X,Koltun V,Guibas L.Joint shape segmentation with linear programming[J].ACM Transactions on Graphics,2011,30(6):Article No.125
[25]Hu R Z,Fan L B,Liu L G.Co-segmentation of 3D shapes via subspace clustering[J].Computer Graphics Forum,2012,31(5):1703-1713
[26]Gatzke T,Grimm C,Garland M,et al.Curvature maps for local shape comparison[C]//Proceedings of International Conference on Shape Modeling and Applications.Los Alamitos:IEEEComputer Society Press,2005:244-253
[27]Vranic D V.An improvement of rotation invariant 3D-shape based on functions on concentric spheres[C]//Proceedings of International Conference on Image Processing.Los Alamitos:IEEE Computer Society Press,2003,3:757-760
[28]Guo K,Zou D G,Chen X W.3D mesh labeling via deep convolutional neural networks[J].ACM Transactions on Graphics,2015,35(1):Article No.3
[29]Ren X,Malik J.Learning a classification model for segmentation[C]//Proceeding of the 9th IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2003,1:10-17
[30]Shi J,Malik J.Normalized cuts and image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8):888-905
[31]LavouéG.Bag of words and local spectral descriptor for 3Dpartial shape retrieval[C]//Proceedings of the 4th Eurographics Conference on 3D Object Retrieval.Aire-la-Ville:Eurographics Association Press,2011:41-48
[32]Bengio Y,Lamblin P,Popovici D,et al.Greedy layer-wise training of deep networks[C]//Proceedings of Annual Conference on Neural Information Processing Systems.Cambridge:MIT Press,2007:153-160
[33]Pérez-Cruz F.Kullback-Leibler divergence estimation of continuous distributions[C]//Proceedings of IEEE International Symposium on Information Theory.Los Alamitos:IEEEComputer Society Press,2008:1666-1670