Learnable Gabor modulated complex-valued networks for orientation robustness
This work provides an incremental improvement for computer vision practitioners seeking to enhance the orientation robustness of CNNs.
The paper addresses the problem of orientation sensitivity in CNNs by modulating complex-valued convolutional weights with learned Gabor filters. This approach allows the network to generate orientation-dependent features without interpolation and offers direct control over equivariance versus invariance. The proposed Learnable Gabor Convolutional Networks (LGCNs) demonstrate rotation invariance and equivariance on MNIST, BSD, and astronomical image datasets.
Robustness to transformation is desirable in many computer vision tasks, given that input data often exhibits pose variance. While translation invariance and equivariance is a documented phenomenon of CNNs, sensitivity to other transformations is typically encouraged through data augmentation. We investigate the modulation of complex valued convolutional weights with learned Gabor filters to enable orientation robustness. The resulting network can generate orientation dependent features free of interpolation with a single set of learnable rotation-governing parameters. By choosing to either retain or pool orientation channels, the choice of equivariance versus invariance can be directly controlled. Moreover, we introduce rotational weight-tying through a proposed cyclic Gabor convolution, further enabling generalisation over rotations. We combine these innovations into Learnable Gabor Convolutional Networks (LGCNs), that are parameter-efficient and offer increased model complexity. We demonstrate their rotation invariance and equivariance on MNIST, BSD and a dataset of simulated and real astronomical images of Galactic cirri.