Semantic enhanced deep learning for image classification

Abstract

Intheabstract,"stacked" should have been "stacked ratio," and "convolution" should have been replaced with "synthetic."-The correct version of FIGURE 1 is presented below: FIGURE 1 Semantic model for image classification -The correct version of FIGURE 2 is presented below: Input Data Image classifier Object layer Ratio Denoising Data Ratio denoising operation Semantic enhancement model based on SrAE Image classification model FIGURE 2The structure of SrAE -In Section 3.1, "denoising" should read "ratio denoising."-"SdAE" should read "SrAE" throughout the paper.-In Section 3.2, "Convolution" should read "Synthetic" in the caption, and "convolution" should read "synthetic" in the body of this section.Concurrency Computat Pract Exper. 2018;30:e4720. wileyonlinelibrary.com/journal/cpe

Using subspaces of weight matrix for evaluating generative adversarial networks with Fréchet distance

“…Recent years have witnessed the breakthrough successes in supervised computer vision tasks such as image classification, 1,2 object detection, 3 and instance segmentation 4 with the help of deep convolutional neural networks (CNNs) 5 . On the other hand, as the counterpart of the discriminative ones, the generative models also have showed exciting performance on creating data, for example, synthesizing diverse images, 6 composing music, 7 face aging, 8 image dehazing 9 which fall into the scope of unsupervised tasks.…”

High performance deep learning techniques for big data analytics

“…A total of 14 features are extracted and further reduced with the Principal Component Analysis (PCA) method. Li et al present a semantic enhanced convolution deep Boltzmann machine that combines the convolutional neural network model with the deep Boltzmann machine for image classification. Two hidden deep learning models are employed to extract image semantics, and the high‐level semantics of an image is obtained by learning from an input image.…”