Learning Separable Filters

By Roberto Rigamonti, Amos Sironi, Vincent Lepetit, Pascal Fua. 2013

308 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition, Efficiency & Compression

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Vision and Pattern Recognition

In Efficiency & Compression

Citations in the corpus

Speeding up Convolutional Neural Networks with Low Rank Expansions (2014, methodology): “One way to exploit this redundancy is to approximate the filter set by a linear combination of a smaller basis set of M filters [32, 35, 36].”
Predicting Parameters in Deep Learning (2013, neutral): “[22] study approximating convolutional filter banks with linear combinations of separable filters.”
Training CNNs with Low-Rank Filters for Efficient Image Classification (2015, neutral): “Rigamonti et al. (2013) approximate previously trained CNNs with low-rank filters for the semantic segmentation of curvilinear structures within volumetric medical imagery.”
Compression of fully-connected layer in neural network by Kronecker product (2015, background): “…at reducing the number of parameters and amount of computation of neural network models, while keeping the quality of prediction results mostly the same.1 Model approximation is important for real world application of neural network to satisfy the time and storage constraints of the applications.”
Compressing Convolutional Neural Networks in the Frequency Domain (2015, methodology): “Two works propose to approximate convolutional filters by a weighted linear combination of basis filters [25, 45].”
Accelerating Very Deep Convolutional Networks for Classification and Detection (2015, methodology): “These methods have shown promising speedup ratios and accuracy on one or two layers and whole (but shallower) models.”
Speeding-up Convolutional Neural Networks Using Fine-tuned CP-Decomposition (2014, methodology): “Using low-rank decomposition to accelerate convolution was suggested by Rigamonti et al. (2013) in the context of codebook learning.”
Deep Roots: Improving CNN Efficiency with Hierarchical Filter Groups (2016, methodology): “filters in the form of low-rank spatial filters [28, 18, 34, 25, 16], using frequencybased convolution [29, 31].”
Flattened Convolutional Neural Networks for Feedforward Acceleration (2014, methodology): “A classic but powerful method to accelerate ﬁltering operations is to condition separability on object function (Rigamonti et al., 2013) so as to force the network to learn separable ﬁlters.”
Sparse 3D convolutional neural networks (2015, background): “Applying the convolutions in Fourier space [11], or using separable filters [13] could help, but simply the amount of memory needed to store large 3D grids of vectors would still be a problem.”
Video Frame Interpolation via Adaptive Separable Convolution (2017, methodology): “Our work is inspired by the success of using separable filters to approximate full 2D filters for other computer vision tasks, like image structure extraction [41].”
Distilling Knowledge From a Deep Pose Regressor Network (2019, methodology): “Most approaches, such as [24, 18, 2], factorize a fully connected or convolutional layer through matrix/tensor decomposition techniques such as SVD or CP-decomposition.”
Compressing deep neural networks using a rank-constrained topology (2015, unknown): “Our approach is similar to previous work on learning separable filters [15, 16] which is used in image processing for speeding up computation in convolution neural networks (see, e.”
A Survey of Model Compression and Acceleration for Deep Neural Networks (2017, unknown): “[35], following the dictionary learning idea.”
A comprehensive survey on model compression and acceleration (2020, methodology): “In the proposed work, to reduce the number of parameters by a factor p , following holds Rigamonti et al. (2013) approximate already trained CNN with low-rank filters.”
Factorized Higher-Order CNNs With an Application to Spatio-Temporal Emotion Estimation (2019, background): “in computer vision by [36] in the context of filter banks.”
Communication-Efficient Edge AI: Algorithms and Systems (2020, background): “filters to accelerate convolution in [82], [83], which is achieved by adding additional regularization to find low-rank filters.”