TT-NF: Tensor Train Neural Fields

Learning neural fields has been an active topic in deep learning research, focusing, among other issues, on finding more compact and easy-to-fit representations. In this paper, we introduce a novel low-rank representation termed Tensor Train Neural Fields (TT-NF) for learning neural fields on dense regular grids and efficient methods for sampling from them. Our representation is a TT parameterization of the neural field, trained with backpropagation to minimize a non-convex objective. We analyze the effect of low-rank compression on the downstream task quality metrics in two settings. First, we demonstrate the efficiency of our method in a sandbox task of tensor denoising, which admits comparison with SVD-based schemes designed to minimize reconstruction error. Furthermore, we apply the proposed approach to Neural Radiance Fields, where the low-rank structure of the field corresponding to the best quality can be discovered only through learning.

[1]  K. Schindler,et al.  tntorch: Tensor Network Learning with PyTorch , 2022, J. Mach. Learn. Res..

[2]  L. Gool,et al.  Pix2NeRF: Unsupervised Conditional $\pi$-GAN for Single Image to Neural Radiance Fields Translation , 2022, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  T. Müller,et al.  Instant neural graphics primitives with a multiresolution hash encoding , 2022, ACM Trans. Graph..

[4]  Benjamin Recht,et al.  Plenoxels: Radiance Fields without Neural Networks , 2021, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  L. Gool,et al.  Implicit Neural Representations for Image Compression , 2021, ECCV.

[6]  Federico Tombari,et al.  Neural Fields in Visual Computing and Beyond , 2021, Comput. Graph. Forum.

[7]  Hwann-Tzong Chen,et al.  Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstruction , 2021, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  Deva Ramanan,et al.  NeRS: Neural Reflectance Surfaces for Sparse-view 3D Reconstruction in the Wild , 2021, NeurIPS.

[9]  Long Quan,et al.  Learning Signed Distance Field for Multi-view Surface Reconstruction , 2021, 2021 IEEE/CVF International Conference on Computer Vision (ICCV).

[10]  Lei Deng,et al.  QTTNet: Quantized tensor train neural networks for 3D object and video recognition , 2021, Neural Networks.

[11]  Andreas Krause,et al.  Cherry-Picking Gradients: Learning Low-Rank Embeddings of Visual Data via Differentiable Cross-Approximation , 2021, 2021 IEEE/CVF International Conference on Computer Vision (ICCV).

[12]  Stefanos Zafeiriou,et al.  Tensor Methods in Computer Vision and Deep Learning , 2021, Proceedings of the IEEE.

[13]  Jonathan T. Barron,et al.  Baking Neural Radiance Fields for Real-Time View Synthesis , 2021, 2021 IEEE/CVF International Conference on Computer Vision (ICCV).

[14]  Pratul P. Srinivasan,et al.  Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fields , 2021, 2021 IEEE/CVF International Conference on Computer Vision (ICCV).

[15]  L. Gool,et al.  Spectral Tensor Train Parameterization of Deep Learning Layers , 2021, AISTATS.

[16]  V. Batista,et al.  Iterative Power Algorithm for Global Optimization with Quantics Tensor Trains. , 2021, Journal of chemical theory and computation.

[17]  I. Oseledets,et al.  TT-TSDF: Memory-Efficient TSDF with Low-Rank Tensor Train Decomposition , 2020, 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[18]  Anru R. Zhang,et al.  Optimal High-Order Tensor SVD via Tensor-Train Orthogonal Iteration , 2020, IEEE Transactions on Information Theory.

[19]  Luc Van Gool,et al.  Reparameterizing Convolutions for Incremental Multi-Task Learning without Task Interference , 2020, ECCV.

[20]  L. Gool,et al.  T-Basis: a Compact Representation for Neural Networks , 2020, ICML.

[21]  Pratul P. Srinivasan,et al.  NeRF , 2020, ECCV.

[22]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[23]  Luc Van Gool,et al.  AI Benchmark: All About Deep Learning on Smartphones in 2019 , 2019, 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW).

[24]  Luc Van Gool,et al.  Learning Filter Basis for Convolutional Neural Network Compression , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[25]  O. Coulaud,et al.  Fast BEM Solution for 2-D Scattering Problems Using Quantized Tensor-Train Format , 2019, IEEE Transactions on Magnetics.

[26]  Maja Pantic,et al.  T-Net: Parametrizing Fully Convolutional Nets With a Single High-Order Tensor , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[27]  Johnnie Gray,et al.  opt\_einsum - A Python package for optimizing contraction order for einsum-like expressions , 2018, J. Open Source Softw..

[28]  Peter Lindstrom,et al.  TTHRESH: Tensor Compression for Multidimensional Visual Data , 2018, IEEE Transactions on Visualization and Computer Graphics.

[29]  Yifan Sun,et al.  Wide Compression: Tensor Ring Nets , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[30]  Alexander Novikov,et al.  Tensor Train decomposition on TensorFlow (T3F) , 2018, J. Mach. Learn. Res..

[31]  Vladimir A. Kazeev,et al.  QTT-finite-element approximation for multiscale problems I: model problems in one dimension , 2017, Adv. Comput. Math..

[32]  Alexander Novikov,et al.  Ultimate tensorization: compressing convolutional and FC layers alike , 2016, ArXiv.

[33]  Maja Pantic,et al.  TensorLy: Tensor Learning in Python , 2016, J. Mach. Learn. Res..

[34]  Anima Anandkumar,et al.  Tensor Contractions with Extended BLAS Kernels on CPU and GPU , 2016, 2016 IEEE 23rd International Conference on High Performance Computing (HiPC).

[35]  Alexander Novikov,et al.  Tensorizing Neural Networks , 2015, NIPS.

[36]  Renato Pajarola,et al.  Analysis of tensor approximation for compression-domain volume visualization , 2015, Comput. Graph..

[37]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[38]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[39]  Ivan V. Oseledets,et al.  Speeding-up Convolutional Neural Networks Using Fine-tuned CP-Decomposition , 2014, ICLR.

[40]  Renato Pajarola,et al.  State‐of‐the‐Art in Compressed GPU‐Based Direct Volume Rendering , 2014, Comput. Graph. Forum.

[41]  Anton Rodomanov,et al.  Putting MRFs on a Tensor Train , 2014, ICML.

[42]  Andrew Zisserman,et al.  Speeding up Convolutional Neural Networks with Low Rank Expansions , 2014, BMVC.

[43]  Boris N. Khoromskij,et al.  Computation of extreme eigenvalues in higher dimensions using block tensor train format , 2013, Comput. Phys. Commun..

[44]  S. V. DOLGOV,et al.  Fast Solution of Parabolic Problems in the Tensor Train/Quantized Tensor Train Format with Initial Application to the Fokker-Planck Equation , 2012, SIAM J. Sci. Comput..

[45]  Reinhold Schneider,et al.  On manifolds of tensors of fixed TT-rank , 2012, Numerische Mathematik.

[46]  Ivan Oseledets,et al.  Tensor-Train Decomposition , 2011, SIAM J. Sci. Comput..

[47]  Eugene E. Tyrtyshnikov,et al.  Algebraic Wavelet Transform via Quantics Tensor Train Decomposition , 2011, SIAM J. Sci. Comput..

[48]  B. Khoromskij O(dlog N)-Quantics Approximation of N-d Tensors in High-Dimensional Numerical Modeling , 2011 .

[49]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[50]  Ivan Oseledets,et al.  Approximation of matrices with logarithmic number of parameters , 2009 .

[51]  Tamara G. Kolda,et al.  Tensor Decompositions and Applications , 2009, SIAM Rev..

[52]  J. Demmel,et al.  An updated set of basic linear algebra subprograms (BLAS) , 2002, TOMS.

[53]  Lawrence D. Jackel,et al.  Backpropagation Applied to Handwritten Zip Code Recognition , 1989, Neural Computation.

[54]  A. Rogozhnikov Einops: Clear and Reliable Tensor Manipulations with Einstein-like Notation , 2022, ICLR.

[55]  E. Tyrtyshnikov,et al.  TT-cross approximation for multidimensional arrays , 2010 .

[56]  S. Goreinov,et al.  The maximum-volume concept in approximation by low-rank matrices , 2001 .