Efficient Training of Convolutional Deep Belief Networks in the Frequency Domain for Application to High-Resolution 2D and 3D Images

57 citations recorded by the source corpus.

How researchers see this paper

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

Compressing Convolutional Neural Networks in the Frequency Domain (2015, background): “Training in the Fourier frequency domain may be advantageous for similar reasons, particularly when convolutions are being performed over large 3-D volumes [4].”
Deep 3D Convolutional Encoder Networks With Shortcuts for Multiscale Feature Integration Applied to Multiple Sclerosis Lesion Segmentation (2016, methodology): “Pre-training and fine-tuning were performed using a highly optimized GPU-accelerated implementation of 3D convRBMs and CENs that performs training in the frequency domain [3].”
Deep learning approach for active classification of electrocardiogram signals (2016, background): “, handcrafted features fed as input to a kernel classifier), deep learning has shown outstanding results in many applications such as image classification [23], object recognition [6], face recognition [29], medical image analysis [10], and time series data [16, 24, 36].”
Using convolutional features and a sparse autoencoder for land-use scene classification (2016, background): “Deep CNNs are composed of several layers of processing, each comprising both linear and non-linear operators, which are learnt jointly, in an end-to-end manner, to solve speci ﬁ c tasks (Farabet et al. 2013; Brosch and Tam 2015).”