Very deep convolutional neural network based image classification using small training sample size

By Shuying Liu, Weihong Deng. 2015

991 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Do ImageNet Classifiers Generalize to ImageNet? (2019, neutral): “vgg_15_BN_64 [40, 49] https://github.”
Do CIFAR-10 Classifiers Generalize to CIFAR-10? (2018, background): “…: Shake-shake [4] with depth 26, base channels 64, S-S-I model, batch size 64, lr 0.1, cosine scheduler, cutout [3] size 16 • vgg16_keras : VGG [14, 18] with depth 16, batch normalization • vgg_15_BN_64 : VGG [14, 18] with depth 15, batch normalization, 64 channels • wide_resnet_tf : Wide…”
Image2StyleGAN: How to Embed Images Into the StyleGAN Latent Space? (2019, background): “[6, 7] observed that the learned filters of the VGG image classification model [20] are excellent general-purpose feature extractors and proposed to use the covariance statistics of the extracted features to measure the high-level similarity between images perceptually, which is then formalized as t”
Selective Classification for Deep Neural Networks (2017, methodology): “We used the VGG-16 architecture [17] and adapted it to the CIFAR-10 dataset by adding massive dropout, exactly as described in [15].”
To Trust Or Not To Trust A Classifier (2018, methodology): “We used a pretrained VGG-16 [49] architecture with adaptation to the CIFAR datasets based on [37].”
LANTERN-RD: Enabling Deep Learning for Mitigation of the Invasive Spotted Lanternfly (2022, methodology): “Alongside the dataset, we train a baseline CNN using the VGG16 [13] model architecture.”
Cross-Modal learning for Audio-Visual Video Parsing (2021, methodology): “For an audio segment, we use the VGGish extractor [3] to get 512-dimensional audio features.”
A Survey of Privacy Attacks in Machine Learning (2020, methodology): “In terms of attack accuracy, they reported up to 90.8% on large neural network models such as VGG16 [71] that were performing classification on the Imagenet [23] dataset.”
Non-Parametric Calibration for Classification (2019, neutral): “The following CNNs were used: AlexNet (Krizhevsky et al., 2012), VGG19 (Simonyan and Zisserman, 2014, Liu and Deng, 2015), ResNet-50, ResNet-152 (He et al., 2016), WideResNet (Zagoruyko and Komodakis, 2016), DenseNet-121, DenseNet-BC-190, DenseNet-201 (Huang et al., 2017), InceptionV4 (Szegedy et…”
Simultaneous extraction of roads and buildings in remote sensing imagery with convolutional neural networks (2017, background): “…deep Convolutional Neural Network (CNN) architectures (Krizhevsky et al., 2012; Lin et al., 2014; Zeiler and Fergus, 2014; Szegedy et al., 2015; Liu and Deng, 2015; Zhou et al., 2014) have been achieving impressive state-of-the art performance for semantic classification, not only in remote…”
SelectiveNet: A Deep Neural Network with an Integrated Reject Option (2019, methodology): “For the convolutional neural network (CNN) experiments, we used the well-known VGG-16 architecture (Simonyan & Zisserman, 2014), optimized for the small datasets and image sizes as suggested by Liu & Deng (2015).”
On The Power of Curriculum Learning in Training Deep Networks (2019, methodology): “Cases 4 and 5 use a public-domain VGG-based architecture5 (Simonyan & Zisserman, 2014; Liu & Deng, 2015), to classify the 3All the code used here will be published upon acceptance.”
Impact of Fully Connected Layers on Performance of Convolutional Neural Networks for Image Classification (2019, methodology): “Another CNN model is designed based on the CIFARVGG [23] model by eliminating some Conv layers from the model.”
Automatic crack classification and segmentation on masonry surfaces using convolutional neural networks and transfer learning (2021, positive): “In particular, the considered networks are: VGG16 [59], MobileNet [60], MobileNetV2 [61], Incep-tionV3 [62], DenseNet121 [63], DenseNet169 [63], ResNet34 [64], ResNet50 [64].”
Within the Lack of Chest COVID-19 X-ray Dataset: A Novel Detection Model Based on GAN and Deep Transfer Learning (2020, methodology): “Many DTL models were introduced like VGGNet [37], GoogleNet [38], ResNet [39], Xception [40], Inception-V3 [41] and DenseNet [42].”
Deep learning for image classification on very small datasets using transfer learning (2019, methodology): “The goal of this work is to show that a very deep convolutonal neural network can be used to fit small dataset like CIFAR-10 with simple but proper modifications [38].”