Abstract
Assembly-based tools provide a powerful modeling paradigm for non-expert shape designers. However, choosing a component from a large shape repository and aligning it to a partial assembly can become a daunting task. In this paper we describe novel neural network architectures for suggesting complementary components and their placement for an incomplete 3D part assembly. Unlike most existing techniques, our networks are trained on unlabeled data obtained from public online repositories, and do not rely on consistent part segmentations or labels. Absence of labels poses a challenge in indexing the database of parts for the retrieval. We address it by jointly training embedding and retrieval networks, where the first indexes parts by mapping them to a low-dimensional feature space, and the second maps partial assemblies to appropriate complements. The combinatorial nature of part arrangements poses another challenge, since the retrieval network is not a function: several complements can be appropriate for the same input. Thus, instead of predicting a single output, we train our network to predict a probability distribution over the space of part embeddings. This allows our method to deal with ambiguities and naturally enables a UI that seamlessly integrates user preferences into the design process. We demonstrate that our method can be used to design complex shapes with minimal or no user input. To evaluate our approach we develop a novel benchmark for component suggestion systems demonstrating significant improvement over state-of-the-art techniques.
Supplemental Material
- Christopher M. Bishop. 1994. Mixture density networks. Technical Report. Neural Computing Research Group, Aston University.Google Scholar
- Angel X. Chang, Thomas A. Funkhouser, Leonidas J. Guibas, Pat Hanrahan, Qi-Xing Huang, Zimo Li, Silvio Savarese, Manolis Sawa, Shuran Song, Hao Su, Jianxiong Xiao, Li Yi, and Fisher Yu. 2015. ShapeNet: An Information-Rich 3D Model Repository. CoRR abs/1512.03012 (2015).Google Scholar
- Siddhartha Chaudhuri, Evangelos Kalogerakis, Leonidas Guibas, and Vladlen Koltun. 2011. Probabilistic Reasoning for Assembly-based 3D Modeling. In SIGGRAPH. Google ScholarDigital Library
- Siddhartha Chaudhuri and Vladlen Koltun. 2010. Data-driven Suggestions for Creativity Support in 3D Modeling. In SIGGRAPH Asia. Google ScholarDigital Library
- Gal Chechik, Varun Sharma, Uri Shalit, and Samy Bengio. 2010. Large Scale Online Learning of Image Similarity Through Ranking. JMLR (2010). Google ScholarDigital Library
- Christopher B. Choy, Danfei Xu, JunYoung Gwak, Kevin Chen, and Silvio Savarese. 2016. 3D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstruction.Google Scholar
- Haoqiang Fan, Hao Su, and Leonidas Guibas. 2017. A Point Set Generation Network for 3D Object Reconstruction from a Single Image. CVPR (2017).Google Scholar
- Matthew Fisher, Daniel Ritchie, Manolis Sawa, Thomas Funkhouser, and Pat Hanrahan. 2012. Example-based Synthesis of 3D Object Arrangements. SIGGRAPH Asia (2012). Google ScholarDigital Library
- Thomas Funkhouser, Michael Kazhdan, Philip Shilane, Patrick Min, William Kiefer, Ayellet Tal, Szymon Rusinkiewicz, and David Dobkin. 2004. Modeling by Example. In SIGGRAPH. Google ScholarDigital Library
- A. Garcia-Garcia, F. Gomez-Donoso, J. Garcia-Rodriguez, S. Orts-Escolano, M. Cazorla, and J. Azorin-Lopez. 2016. PointNet: A 3D Convolutional Neural Network for real-time object class recognition. In IJCNN.Google Scholar
- Rohit Girdhar, David F. Fouhey, Mikel Rodriguez, and Abhinav Gupta. 2016. Learning a Predictable and Generative Vector Representation for Objects.Google Scholar
- Aleksey Golovinskiy and Thomas Funkhouser. 2008. Randomized Cuts for 3D Mesh Analysis. ACM Transactions on Graphics (Proc. SIGGRAPH ASIA) 27, 5 (Dec. 2008). Google ScholarDigital Library
- Aleksey Golovinskiy and Thomas Funkhouser. 2009. Consistent Segmentation of 3D Models. Proc. SMI (2009).Google ScholarDigital Library
- Edward Grant, Pushmeet Kohli, and Marcel van Gerven. 2016. Deep Disentangled Representations for Volumetric Reconstruction.Google Scholar
- R. Hadsell, S. Chopra, and Y. LeCun. 2006. Dimensionality Reduction by Learning an Invariant Mapping. In CVPR. Google ScholarDigital Library
- R. Hu, L. Fan, and L. Liu. 2012. Co-segmentation of 3D shapes via subspace clustering. SGP (2012). Google ScholarDigital Library
- Ruizhen Hu, Chenyang Zhu, Oliver van Kaick, Ligang Liu, Ariel Shamir, and Hao Zhang. 2015. Interaction Context (ICON): Towards a Geometric Functionality Descriptor. SIGGRAPH (2015). Google ScholarDigital Library
- Qixing Huang, Vladlen Koltun, and Leonidas Guibas. 2011. Joint shape segmentation with linear programming. In SIGGRAPH Asia. Google ScholarDigital Library
- Qixing Huang, Fan Wang, and Leonidas Guibas. 2014. Functional Map Networks for Analyzing and Exploring Large Shape Collections. ACM TOG 33 (2014). Google ScholarDigital Library
- Prakhar Jaiswal, Jinmiao Huang, and Rahul Rai. 2016. Assembly-based conceptual 3D modeling with unlabeled components using probabilistic factor graph. Computer-Aided Design (2016). Google ScholarDigital Library
- Evangelos Kalogerakis, Siddhartha Chaudhuri, Daphne Koller, and Vladlen Koltun. 2012. A Probabilistic Model for Component-based Shape Synthesis. SIGGRAPH (2012). Google ScholarDigital Library
- Evangelos Kalogerakis, Aaron Hertzmann, and Karan Singh. 2010. Learning 3D mesh segmentation and labeling. In SIGGRAPH. Google ScholarDigital Library
- Vladimir G. Kim, Siddhartha Chaudhuri, Leonidas Guibas, and Thomas Funkhouser. 2014. Shape2Pose: Human-centric Shape Analysis. SIGGRAPH (2014). Google ScholarDigital Library
- Vladimir G. Kim, Wilmot Li, Niloy J. Mitra, Siddhartha Chaudhuri, Stephen DiVerdi, and Thomas Funkhouser. 2013. Learning Part-based Templates from Large Collections of 3D Shapes. SIGGRAPH (2013). Google ScholarDigital Library
- Diederik P. Kingma and Jimmy Ba. 2014. Adam: A Method for Stochastic Optimization. CoRR (2014).Google Scholar
- Diederik P. Kingma, Danilo J. Rezende, Shakir Mohamed, and Max Welling. 2014. Semi-supervised Learning with Deep Generative Models. In NIPS. Google ScholarDigital Library
- Chen Kong, Chen-Hsuan Lin, and Simon Lucey. 2017. Using Locally Corresponding CAD Models for Dense 3D Reconstructions from a Single Image. CVPR (2017).Google Scholar
- Jun Li, Kai Xu, Siddhartha Chaudhuri, Ersin Yumer, Hao Zhang, and Leonidas Guibas. 2017. GRASS: Generative Recursive Autoencoders for Shape Structures. SIGGRAPH (2017). Google ScholarDigital Library
- Yangyan Li, Angela Dai, Leonidas Guibas, and Matthias Nießner. 2015. Database-Assisted Object Retrieval for Real-Time 3D Reconstruction. (2015).Google Scholar
- A. Makadia and M. E. Yumer. 2014. Learning 3D Part Detection from Sparsely Labeled Data. In 3DV. Google ScholarDigital Library
- Mehdi Mirza and Simon Osindero. 2014. Conditional Generative Adversarial Nets. CoRR abs/1411.1784 (2014).Google Scholar
- Robert Osada, Thomas Funkhouser, Bernard Chazelle, and David Dobkin. 2002. Shape Distributions. ACM TOG (2002). Google ScholarDigital Library
- Charles Ruizhongtai Qi, Hao Su, Kaichun Mo, and Leonidas J. Guibas. 2017. PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation. CVPR (2017).Google Scholar
- C. R. Qi, H. Su, M. Nieçner, A. Dai, M. Yan, and L. J. Guibas. 2016. Volumetric and Multi-view CNNs for Object Classification on 3D Data. In CVPR.Google Scholar
- Gernot Riegler, Ali Osman Ulusoy, and Andreas Geiger. 2017. OctNet: Learning Deep 3D Representations at High Resolutions. CVPR (2017).Google Scholar
- Tianjia Shao, Aron Monszpart, Youyi Zheng, Bongjin Koo, Weiwei Xu, Kun Zhou, and Niloy J. Mitra. 2014. Imagining the Unseen: Stability-based Cuboid Arrangements for Scene Understanding. SIGGRAPH Asia (2014). Google ScholarDigital Library
- Abhishek Sharma, Oliver Grau, and Mario Fritz. 2016. VConv-DAE: Deep Volumetric Shape Learning Without Object Labels.Google Scholar
- Chao-Hui Shen, Hongbo Fu, Kang Chen, and Shi-Min Hu. 2012. Structure Recovery by Part Assembly. SIGGRAPH Asia (2012). Google ScholarDigital Library
- Oana Sidi, Oliver van Kaick, Yanir Kleiman, Hao Zhang, and Daniel Cohen-Or. 2011. Unsupervised Co-Segmentation of a Set of Shapes via Descriptor-Space Spectral Clustering. SIGGRAPH Asia (2011). Google ScholarDigital Library
- Kihyuk Sohn, Xinchen Yan, and Honglak Lee. 2015. Learning Structured Output Representation Using Deep Conditional Generative Models. In NIPS. Google ScholarDigital Library
- Hao Su, Yangyan Li, Charles Ruizhongtai Qi, Noa Fish, Daniel Cohen-Or, and Leonidas J Guibas. 2015a. Joint embeddings of shapes and images via cnn image purification. ACM Transactions on Graphics (TOG) 34, 6 (2015), 234. Google ScholarDigital Library
- Hang Su, Subhransu Maji, Evangelos Kalogerakis, and Erik Learned-Miller. 2015b. Multi-view Convolutional Neural Networks for 3D Shape Recognition. In ICCV. Google ScholarDigital Library
- Minhyuk Sung, Vladimir G. Kim, Roland Angst, and Leonidas Guibas. 2015. Data-driven Structural Priors for Shape Completion. SIGGRAPH Asia (2015). Google ScholarDigital Library
- Trimble. 2017. 3D Warehouse. (2017). https://3dwarehouse.sketchup.com/Google Scholar
- Shubham Tulsiani, Tinghui Zhou, Alexei A. Efros, and Jitendra Malik. 2017. Multi-view Supervision for Single-view Reconstruction via Differentiable Ray Consistency. In CVPR.Google Scholar
- Laurens van der Maaten. 2009. Learning a Parametric Embedding by Preserving Local Structure. In AISTATS.Google Scholar
- Oliver van Kaick, Kai Xu, Hao Zhang, Yanzhen Wang, Shuyang Sun, Ariel Shamir, and Daniel Cohen-Or. 2013. Co-hierarchical analysis of shape structures. ACM TOG (2013). Google ScholarDigital Library
- Yunhai Wang, Shmulik Asafi, Oliver van Kaick, Hao Zhang, Daniel Cohen-Or, and Baoquan Chen. 2012. Active Co-analysis of a Set of Shapes. SIGGRAPH Asia (2012). Google ScholarDigital Library
- Jiajun Wu, Tianfan Xue, Joseph J. Lim, Yuandong Tian, Joshua B. Tenenbaum, Antonio Torralba, and William T. Freeman. 2016a. Single Image 3D Interpreter Network.Google Scholar
- Jiajun Wu, Chengkai Zhang, Tianfan Xue, William T Freeman, and Joshua B Tenenbaum. 2016b. Learning a probabilistic latent space of object shapes via 3d generative-adversarial modeling. In NIPS. Google ScholarDigital Library
- Zhirong Wu, Shuran Song, Aditya Khosla, Xiaoou Tang, and Jianxiong Xiao. 2015a. 3D ShapeNets for 2.5D Object Recognition and Next-Best-View Prediction. CVPR (2015).Google Scholar
- Zhirong Wu, S. Song, A. Khosla, Fisher Yu, Linguang Zhang, Xiaoou Tang, and J. Xiao. 2015b. 3D ShapeNets: A deep representation for volumetric shapes. In CVPR.Google Scholar
- Zhige Xie, Kai Xu, Ligang Liu, and Yueshan Xiong. 2014. 3D Shape Segmentation and Labeling via Extreme Learning Machine. SGP (2014). Google ScholarDigital Library
- Kai Xu, Hao Zhang, Daniel Cohen-Or, and Baoquan Chen. 2012. Fit and Diverse: Set Evolution for Inspiring 3D Shape Galleries. SIGGRAPH (2012). Google ScholarDigital Library
- Li Yi, Leonidas Guibas, Aaron Hertzmann, Vladimir G. Kim, Hao Su, and Ersin Yumer. 2017. Learning Hierarchical Shape Segmentation and Labeling from Online Repositories. SIGGRAPH (2017). Google ScholarDigital Library
- Li Yi, Vladimir G. Kim, Duygu Ceylan, I-Chao Shen, Mengyan Yan, Hao Su, Cewu Lu, Qixing Huang, Alla Sheffer, and Leonidas Guibas. 2016. A Scalable Active Framework for Region Annotation in 3D Shape Collections. SIGGRAPH Asia (2016). Google ScholarDigital Library
- Youyi Zheng, Daniel Cohen-Or, Melinos Averkiou, and Niloy J. Mitra. 2014. Recurring part arrangements in shape collections. (2014).Google Scholar
- Youyi Zheng, Daniel Cohen-Or, and Niloy J. Mitra. 2013. Smart Variations: Functional Substructures for Part Compatibility. Eurographics (2013).Google Scholar
Index Terms
- ComplementMe: weakly-supervised component suggestions for 3D modeling
Recommendations
Spectral gradient fields embedding for nonrigid shape matching
Display Omitted An embedding of 3D-shapes using spectral gradient fields (GFs).It is based on the inner and external products of pairs of eigenfunctions.The embedding naturally defines a distance measure between nonrigid shapes.The method achieves state-...
Unify 3D Shape Retrieval and Classification in One Framework
DMCIT '17: Proceedings of the 2017 International Conference on Data Mining, Communications and Information TechnologyIn conventional 3D shape retrieval and classification, they both need to compute the similarity between the input shapes and candidate or training 3D shapes. As for 3D shape classification, it essentially computes the distance between the query shapes ...
A 3D shape descriptor based on spectral analysis of medial axis
The medial axis of a 3D shape is widely known for its ability as a compact and complete shape representation. However, there is still lack of a generative description defined over the medial axis directly which limits its actual application to 3D shape ...
Comments