Learning rotation invariant convolutional filters for texture classification

By Diego Marcos, Michele Volpi, Devis Tuia. 2016

153 citations recorded by the source corpus.

We present a method for learning discriminative filters using a shallow Convolutional Neural Network (CNN). We encode rotation invariance directly in the model by tying the weights of groups of filters to several rotated versions of the canonical filter in the group. These filters can be used to extract rotation invariant features well-suited for image classification. We test this learning procedure on a texture classification benchmark, where the orientations of the training images differ from those of the test images. We obtain results comparable to the state-of-the-art. Compared to standard shallow CNNs, the proposed method obtains higher classification performance while reducing by an order of magnitude the number of parameters to be learned.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

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

In Computer Vision and Pattern Recognition

Citations in the corpus

General E(2)-Equivariant Steerable CNNs (2019, neutral): “04 16 regular/scalar ψ0 conv −−→ ρreg G-pool −−−−→ ψ0 ELU, G-pooling [6, 36] 2.”
RotDCF: Decomposition of Convolutional Filters for Rotation-Equivariant Deep Networks (2018, methodology): “[8, 38] proposed rotating conventional ﬁlters to perform rotation-invariant texture and image classiﬁcation.”
Harmonic Networks: Deep Translation and Rotation Equivariance (2016, methodology): “[24] use a larger number of rotations for texture classification and [26] also use many rotated handcrafted filter copies, opting not to learn the filters.”
Learning Steerable Filters for Rotation Equivariant CNNs (2017, background): “Marcos et al. [21] perform convolutions with rotated versions of a each ﬁlter in a shallow network followed by a global pooling over orientations.”
CubeNet: Equivariance to 3D Rotation and Translation (2018, background): “Recently there has been an explosion of interest into CNNs with predefined transformation equivariances, beyond translation [8, 50, 11, 29, 16, 42, 36, 14, 19, 31, 18, 22, 15, 55, 49, 26, 25, 48, 33, 28, 9].”
Oriented Response Networks (2017, background): “…rotating conventional ﬁlters to perform rotation-invariant texture and image classiﬁcation; however, without upgrading conventional ﬁlters to multi-oriented ﬁlters with orientation channels, their capability about capturing hierarchical and ﬁne-detailed orientation information remains limited.”
Graph-based Isometry Invariant Representation Learning (2017, methodology): “Then, the work in (Marcos et al., 2016) suggests using rotated filter banks and a special max pooling operation to combine their outcomes and improve invariance to transformations.”
Rotation-Sensitive Regression for Oriented Scene Text Detection (2018, background): “Some recent works [10, 7, 52] focus on enhancing the rotation invariance of the CNN features to further improve the classification performance.”
Tree Species Identification from Bark Images Using Convolutional Neural Networks (2018, unknown): “Also, [25] modified the standard convolutional layer to learn rotation-invariant filters.”
Deep Rotation Equivariant Network (2017, background): “Methods proposed by [7], [8] show that rotation convolution layers followed by a cross-channel pooling over rotations could achieve rotation equivariance.”