Abstract
Deep learning has recently emerged as one of the most popular and powerful paradigms for learning tasks. In this paper, we present a deep learning approach to 3D shape classification using convolutional neural networks. The proposed framework takes a multi-stage approach that first represents each 3D shape in the dataset as a 2D image using the bag-of-features model in conjunction with intrinsic spatial pyramid matching that leverages the spatial relationship between features. These 2D images are then fed into a pre-trained convolutional neural network to learn deep convolutional shape-aware descriptors from the penultimate fully-connected layer of the network. Finally, a multiclass support vector machine classifier is trained on the deep descriptors, and the classification accuracy is subsequently computed. The effectiveness of our approach is demonstrated on three standard 3D shape benchmarks, yielding higher classification accuracy rates compared to existing methods.
Similar content being viewed by others
References
Goodfellow I, Bengio Y, Courville A (2016) Deep learning. The MIT Press, Cambridge
LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521(7553):436–444
Schmidhuber J (2015) Deep learning in neural networks: an overview. Neural Netw 61:85–117
Su H, Maji S, Kalogerakis E, Learned-Miller E (2015) Multi-view convolutional neural networks for 3D shape recognition. In: Proceedings of ICCV, pp 945–953
Qi C, Su H, Nießner M, Dai A, Yan M, Guibas L (2016) Volumetric and multi-view CNNs for object classification on 3D data. In: Proceedings of CVPR
Wu J, Zhang C, Xue T, Freeman B, Tenenbaum J (2016) Learning a probabilistic latent space of object shapes via 3D generative-adversarial modeling. In: NIPS
Rustamov R (2007) Laplace–Beltrami eigenfunctions for deformation invariant shape representation. In: Proceedings of symposium geometry processing, pp 225–233
Sun J, Ovsjanikov M, Guibas L (2009) A concise and provably informative multi-scale signature based on heat diffusion. Comput Graph Forum 28(5):1383–1392
Bronstein M, Kokkinos I (2010) Scale-invariant heat kernel signatures for non-rigid shape recognition. In: Proceedings of CVPR, pp 1704–1711
Aubry M, Schlickewei U, Cremers D (2011) The wave kernel signature: a quantum mechanical approach to shape analysis. In: Proceedings of computational methods for the innovative design of electrical devices, pp 1626–1633
Li C, Ben Hamza A (2013) A multiresolution descriptor for deformable 3D shape retrieval. Vis Comput 29:513–524
Guler R, Tari S, Unal G (2014) Screened poisson hyperfields for shape coding. SIAM J Imaging Sci 7(4):2558–2590
Ye J, Yu Y (2015) A fast modal space transform for robust nonrigid shape retrieval. Vis Comput 32(5):553–568
Rosenberg S (1997) The Laplacian on a Riemannian manifold. Cambridge University Press, Cambridge
Bronstein A, Bronstein M, Kimmel R (2008) Numerical geometry of non-rigid shapes. Springer, New York
Krim H, Ben Hamza A (2015) Geometric methods in signal and image analysis. Cambridge University Press, Cambridge
Reuter M, Wolter F, Peinecke N (2006) Laplace–Beltrami spectra as ‘Shape-DNA’ of surfaces and solids. Comput Aided Des 38(4):342–366
Ben Hamza A (2016) A graph-theoretic approach to 3D shape classification. Neurocomputing 211:11–21
Nowak E, Jurie F, Triggs B (2006) Sampling strategies for bag-of-features image classification. In: Proceedings of ECCV
Bronstein A, Bronstein M, Guibas L, Ovsjanikov M (2011) Shape Google: geometric words and expressions for invariant shape retrieval. ACM Trans Graph 30(1):1
Litman R, Bronstein A, Bronstein M, Castellani U (2014) Supervised learning of bag-of-features shape descriptors using sparse coding. Comput Graph Forum 33(5):127–136
Bu S, Liu Z, Han J, Wu J, Ji R (2014) Learning high-level feature by deep belief networks for 3-D model retrieval and recognition. IEEE Trans Multimed 24(16):2154–2167
Lipman Y, Rustamov R, Funkhouser T (2010) Biharmonic distance. ACM Trans Graph 29(3):1–11
Wu Z, Song S, Khosla A, Yu F, Zhang L, Tang X, Xiao J (2015) 3D ShapeNets: a deep representation for volumetric shapes. In: Proceedings of CVPR, pp 1912–1920
Zhu Z, Wang X, Bai S, Yao C, Bai X (2016) Deep learning representation using autoencoder for 3D shape retrieval. Neurocomputing 204:41–50
Bronstein M, Bruna J, LeCun Y, Szlam A, Vandergheynst P (2016) Geometric deep learning: going beyond Euclidean data. arXiv:1611.08097
Krizhevsky A, Sutskever I, Hinton G (2012) ImageNet classification with deep convolutional neural networks. In: NIPS pp 1097–1105
Simonyan K, Zisserman A (2015) Very deep convolutional networks for large-scale image recognition. In: ICLR
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of CVPR
Kanezaki A, Matsushita Y, Nishida Y (2016) RotationNet: joint learning of object classification and viewpoint estimation using unaligned 3D object dataset. arXiv:1603.06208
Bu S, Wanga L, Hana P, Liu Z, Li K (2017) 3-D shape recognition and retrieval based on multi-modality deep learning. Neurocomputing 259:183–193
Wang J, Yang J, Yu K, Lv F, Huang T, Gong Y (2010) Locality-constrained linear coding for image classification. In: Proceedings of CVPR, pp 3360–3367
Li C, Ben Hamza A (2013) Intrinsic spatial pyramid matching for deformable 3D shape retrieval. Int J Multimed Inf Retr 2:261–271
LeCun Y, Boser B, Denker J, Henderson D, Howard R, Hubbard W, Jackel L (1989) Backpropagation applied to handwritten zip code recognition. Neural Comput 1(4):541–551
Chatfield K, Simonyan K, Vedaldi A, Zisserman A (2014) Return of the devil in the details: delving deep into convolutional nets. In: Proceedings of BMVC
Lian Z, Zhang J, Choi S, ElNaghy H, El-Sana J, Furuya T, GiachettiA, Isaia RGL, Lai L, Li C, Li H, Limberger F, Martin R, Nakanishi R, Nonato ANL, Ohbuchi R, Pevzner K, Pickup D, Rosin P, Sharf A, Sun L, Sun X, Tari S, Unal G, Wilson R (2015) SHREC’15 track: non-rigid3D shape retrieval. In: Proceedings of eurographics workshop on 3D object retrieval, pp 1–14
Shi B, Bai S, Zhou Z, Bai X (2015) Deeppano: deep panoramic representation for 3D shape recognition. IEEE Signal Process Lett 22(12):2339–2343
Sinha A, Bai J, Ramani K (2016) Deep learning 3D shape surfaces using geometry images. In: Proceedings of ECCV
Qi C, Yi L, Su H, Guibas L (2017) Pointnet++: deep hierarchical feature learning on point sets in a metric space. In: NIPS
van der Maaten L, Hinton G (2008) Visualizing data using t-SNE. J Mach Learn Res 9:2579–2605
Acknowledgements
This work was supported in part by NSERC Discovery Grant N00929.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Ghodrati, H., Luciano, L. & Hamza, A.B. Convolutional Shape-Aware Representation for 3D Object Classification. Neural Process Lett 49, 797–817 (2019). https://doi.org/10.1007/s11063-018-9858-9
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11063-018-9858-9