基于形状先验的隐式三维建模算法研究
Research on Implicit 3D Modeling Algorithm Based on Shape Prior
DOI: 10.12677/CSA.2023.1311199, PDF,  被引量    国家自然科学基金支持
作者: 倪天杰, 何良华：同济大学计算机科学与技术系，上海
关键词: 隐式神经场；深度学习；三维建模；点云卷积；符号距离函数；Implicit Neural Field； Deep Learning； 3D Modeling； Point Cloud Convolution； Signal Distance Function
摘要: 近年来，隐式神经三维建模算法是计算机视觉领域的热门研究方向，特别是基于符号距离函数(SDF)的方法，然而目前的模型往往泛化性不强。由此问题，该文利用动态图卷积网络对点云进行形状编码，引入了形状先验假设，提出了一种结合物体个性特征和其所属类别共性信息相结合的隐式三维建模算法，提升了模型的准确性与泛化性。在ShapeNetV2数据集上，相较于现有算法取得了更好的效果，表明了本方法在三维隐式建模问题上的优越性。
Abstract: In recent years, implicit neural 3D modeling algorithm has become a popular research direction in the field of computer vision, especially the method based on signal distance function (SDF). However, this kind of model is often not strong generalization. To solve this problem, this paper uses the dynamic graph convolutional network to encode the shape of objects, introduces the shape prior hy-pothesis, and proposes an implicit 3D modeling algorithm that combines the individual characteristics of objects and the generic information of their categories, which improves the accuracy and generalization of the model. On ShapeNetV2 dataset, better results are obtained compared with the existing algorithms, which shows the superiority of the proposed method in three-dimensional implicit modeling.
文章引用：倪天杰, 何良华. 基于形状先验的隐式三维建模算法研究[J]. 计算机科学与应用, 2023, 13(11): 2012-2021. https://doi.org/10.12677/CSA.2023.1311199

参考文献

[1] Chen, Z.Q. and Hao, Z. (2019) Learning Implicit Fields for Generative Shape Modeling. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, 15-20 June 2019, 5939-5948. [Google Scholar] [CrossRef] 
[2] Mildenhall, B., Srinivasan, P.P., Tancik, M., Barron, J.T., et al. (2020) NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, JM., Eds., Computer Vision—ECCV 2020. Lecture Notes in Computer Science, Springer, Cham, 405-421. [Google Scholar] [CrossRef] 
[3] Wang, P., Liu, L.J., Liu, Y., Theobalt, C., Komura, T. and Wang, W.P. (2021) NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-View Reconstruction. 35th Conference on Neural Information Processing Systems (NeurlIPS), 6-14 December 2021, 123-131.
[4] Park, J.J., Florence, P., Straub, J., Newcombe, R. and Lovergrove, S. (2019) Deepsdf: Learning Continuous Signed Distance Func-tions for Shape Representation. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, 15-20 June 2019, 165-174. [Google Scholar] [CrossRef] 
[5] Groueix, T., Fisher, M., Kim, V.G., Russell, B.C. and Aubry, M. (2018) A Papier-Mache Approach to Learning 3D Surface Generation. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, 18-23 June 2018, 216-214. [Google Scholar] [CrossRef] 
[6] Mescheder, L., Oechsle, M., Niemeyer, M., Nowozin, S. and Gei-ger, A. (2019) Occupancy Networks: Learning 3D Reconstruction in Function Space. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, 15-20 June 2019, 4455-4465. [Google Scholar] [CrossRef] 
[7] Jiang, C., Sud, A., Makadia, A., Huang, J., Nießner, M. and Funkhouser, T. (2020) Local Implicit Grid Representations for 3D Scenes. 2020 IEEE/CVF Conference on Computer Vi-sion and Pattern Recognition (CVPR), Seattle, 13-19 June 2020, 6000-6009. [Google Scholar] [CrossRef] 
[8] Charles, R.Q., Su, H., Kaichun, M. and Guibas, L.J. (2017) PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation. 2017 IEEE/CVF Conference on Com-puter Vision and Pattern Recognition (CVPR), Honolulu, 21-26 July 2017, 77-85. [Google Scholar] [CrossRef] 
[9] Qi, C.R., Yi, L., Su, H. and Guibas, L.J. (2019) Pointnet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space. 2017 31st Conference on Neural Information Processing Systems (NeurlIPS), Long Beach, 4-9 December 2017, 5105-5114.
[10] Wang, Y., Sun, Y.b., Liu, Z.w., Sarma, S.E., Bronstein, M.M. and Solomon, J.M. (2019) Dynamic Graph CNN for Learning on Point Clouds. ACM Transactions on Graphics (TOG), 38, 1-12. [Google Scholar] [CrossRef] 
[11] Newcombe, R.A., et al. (2011) Kinect Fusion: Real-Time Dense Surface Mapping and Tracking. 2011 10th IEEE International Symposium on Mixed and Augmented Reality (ISMAR), Basel, 26-29 October 2011, 127-136. [Google Scholar] [CrossRef] 
[12] Lorensen, W.E. and Cline, H.E. (1987) Marching Cubes: A High Resolution 3d Surface Construction Algorithm. ACM SIGGRAPH Compuer Graphics, 21, 163-169. [Google Scholar] [CrossRef] 
[13] Bao, S.Y., Chandraker, M., Lin Y. and Savarese, S. (2013) Dense Ob-ject Reconstruction with Semantic Priors. 2013 IEEE Conference on Computer Vision and Pattern Recognition, Portland, 23-28 June 2013, 1264-1271. [Google Scholar] [CrossRef] 
[14] Tancik, M., et al. (2021) Learned Initializations for Optimizing Coor-dinate-Based Neural Representations. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, 20-25 June 2021, 2845-2854. [Google Scholar] [CrossRef] 
[15] Vincent, S., et al. (2020) Implicit Neural Representations with Periodic Activation Functions. 2020 34th Conference on Neural Information Processing Systems (NeurlIPS), 4-9 December 2020, 7462-7473.