ImageNet classification with deep convolutional neural networks

@article{Krizhevsky2012ImageNetCW,
  title={ImageNet classification with deep convolutional neural networks},
  author={Alex Krizhevsky and Ilya Sutskever and Geoffrey E. Hinton},
  journal={Communications of the ACM},
  year={2012},
  volume={60},
  pages={84 - 90}
}
We trained a large, deep convolutional neural network to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes. [] Key Method The neural network, which has 60 million parameters and 650,000 neurons, consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax.
Deep Convolutional Networks without Learning the Classifier Layer
TLDR
This research examines this idea, extends it further and demonstrates that fixed classifiers offer no additional benefit compared to simply removing the output layer along with its parameters, and reveals that the typical approach of having a fully connected final output layer is inefficient in terms of parameter.
Shallow convolutional neural network for image classification
TLDR
A novel shallow convolutional neural network (SCNNB) is proposed to overcome the above limitations for image classification, which uses batch normalization techniques to accelerate training convergence and improve the accuracy.
Shallow convolutional neural network for image classification
  • Dai
  • Computer Science
  • 2019
TLDR
A novel shallow convolutional neural network (SCNNB) is proposed to overcome the above limitations for image classification, which uses batch normalization techniques to accelerate training convergence and improve the accuracy.
Tiny ImageNet Classification with Convolutional Neural Networks
TLDR
This work trained several deep convolutional neural networks to classify 10,000 images from the Tiny ImageNet dataset into 200 distinct classes and found the top single model achieves an error rate of 58.2%.
Object Classification using Deep Convolutional Neural Networks
TLDR
The objective is to determine the best way of using techniques like max pooling when designing a deep neural network architecture for classifying objects, and to develop a framework for optimizing a deep Neural Network Architecture for object classification.
Deeper into Image Classification
TLDR
Inspired by the performance of deep learning models in image classification, the present paper proposed three techniques and implemented that for image classification: residual network, convolutional neural network, and logistic regression were used for classification.
Understanding of a convolutional neural network
TLDR
All the elements and important issues related to CNN, and how these elements work, are explained and defined and the parameters that effect CNN efficiency are state.
Image Classification Based on the Boost Convolutional Neural Network
TLDR
This paper presents the usage of a trained deep convolutional neural network model to extract the features of the images, and then, used the AdaBoost algorithm to assemble the Softmax classifiers into recognizable images, resulting in a 3% increase of accuracy of the trained CNN models.
An Efficient Deep Convolutional Neural Network for Visual Image Classification
TLDR
The proposed model utilizes the well-known convolutional neural network composites to construct a robust Visual Image Classification Network (VICNet), which has a high potential capability in the recognition of test images.
...
...

References

SHOWING 1-10 OF 50 REFERENCES
ImageNet Classification with Deep Convolutional Neural
  • Computer Science, Environmental Science
  • 2013
The intended goal of the experiments was to create a deep, convolutional network that uses supervised learning to achieve better (lower) error rates than the rates previously observed, to identify
Convolutional Deep Belief Networks on CIFAR-10
TLDR
Using a combination of locally-connected convolutional units and globally-connected units, as well as a few tricks to reduce the e ects of over tting, the DBN achieves state-of-the-art performance in the classi cation task of the CIFAR-10 subset of the tiny images dataset.
High-Performance Neural Networks for Visual Object Classification
We present a fast, fully parameterizable GPU implementation of Convolutional Neural Network variants. Our feature extractors are neither carefully designed nor pre-wired, but rather learned in a
Multi-column deep neural networks for image classification
TLDR
On the very competitive MNIST handwriting benchmark, this method is the first to achieve near-human performance and improves the state-of-the-art on a plethora of common image classification benchmarks.
Learning Multiple Layers of Features from Tiny Images
TLDR
It is shown how to train a multi-layer generative model that learns to extract meaningful features which resemble those found in the human visual cortex, using a novel parallelization algorithm to distribute the work among multiple machines connected on a network.
Convolutional networks and applications in vision
TLDR
New unsupervised learning algorithms, and new non-linear stages that allow ConvNets to be trained with very few labeled samples are described, including one for visual object recognition and vision navigation for off-road mobile robots.
Metric Learning for Large Scale Image Classification: Generalizing to New Classes at Near-Zero Cost
TLDR
The goal is to devise classifiers which can incorporate images and classes on-the-fly at (near) zero cost and to explore k-nearest neighbor (k-NN) and nearest class mean (NCM) classifiers.
Best practices for convolutional neural networks applied to visual document analysis
TLDR
A set of concrete bestpractices that document analysis researchers can use to get good results with neural networks, including a simple "do-it-yourself" implementation of convolution with a flexible architecture suitable for many visual document problems.
Learning methods for generic object recognition with invariance to pose and lighting
  • Yann LeCun, F. Huang, L. Bottou
  • Computer Science
    Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004.
  • 2004
TLDR
A real-time version of the system was implemented that can detect and classify objects in natural scenes at around 10 frames per second and proved impractical, while convolutional nets yielded 16/7% error.
...
...