Visualizing and Understanding Convolutional Neural Networks

By Matthew D. Zeiler, Rob Fergus. 2013

556 citations recorded by the source corpus.

Subfields: 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 Computer Vision and Pattern Recognition

Citations in the corpus

Deep Residual Learning for Image Recognition (2015, background): “present a residual learning framework to ease the training of networks that are substantially deeper than those used previously.”
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (2015, methodology): “trained network, we use the “fast” version of ZF net [32] that has five convolutional layers and three fully-connected layers, and the public VGG-16 model(8) [3] that has 13 convolutional layers and three fully-connected layers.”
Object Detection Networks on Convolutional Feature Maps (2015, methodology): “131 This logical division naturally applies to a series of 132 networks, such as AlexNet [5], Zeiler and Fergus’s 133 (ZF) net [25], OverFeat [22], and VGG nets [15], 134 which all have multiple fine-tunable fc layers.”
Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification (2015, background): “Despite the prevalence of rectifier networks, recent improvements of models [37, 28, 12, 29, 33] and theoretical guidelines for training them [8, 27] have rarely focused on the properties of the rectifiers.”
Convolutional neural networks at constrained time cost (2014, methodology): “Here “AlexNet” is our re-implementation of [14], except that we ignore the two-GPU splitting; ZF (fast) is our re-implementation of the smaller model in Zeiler and Fergus’s [24]; CNN-F is the fast model in Chatﬁeld et al .’s [1], which has a structure similar to AlexNet but has fewer ﬁlters; SPPnet ”
Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition (2014, methodology): “This method avoids repeatedly computing the convolutional features.”
Intriguing properties of neural networks (2013, background): “Already, a variety of recent state of the art computer vision models employ input deformations during training for increasing the robustness and convergence speed of the models [9, 13].”
Rethinking ImageNet Pre-Training (2018, background): “Deep convolutional neural networks [21, 23] revolutionized computer vision arguably due to the discovery that feature representations learned on a pre-training task can transfer useful information to target tasks [9, 6, 50].”
Exploring the Limits of Weakly Supervised Pretraining (2018, background): “This formula [1,2,3] has been in wide use for several years and led to impressive improvements on numerous tasks.”
Group Normalization (2018, methodology): “Local Response Normalization (LRN) [26,27,28] was a component in AlexNet [28] and following models [29,30,31].”
Deep Fragment Embeddings for Bidirectional Image Sentence Mapping (2014, background): “In Computer Vision, Convolutional Neural Networks (CNNs) [23] have recently been shown to learn powerful image representations that support state of the art image classification [24, 25, 26] and object detection [27, 28].”
Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour (2017, background): “Moreover, this pattern generalizes: larger datasets and network architectures consistently yield improved accuracy across all tasks that beneﬁt from pre-training [22, 40, 33, 34, 35, 16].”
Aggregated Residual Transformations for Deep Neural Networks (2016, background): “Research on visual recognition is undergoing a transition from “feature engineering” to “network engineering” [25, 24, 44, 34, 36, 38, 14].”
Instance-Aware Semantic Segmentation via Multi-task Network Cascades (2015, methodology): “We show results using ZF net [30] that has 5 convolutional layers and 3 fc layers, and VGG-16 net [27] that has 13 convolutional layers and 3 fc layers.”
Convolutional feature masking for joint object and stuff segmentation (2014, methodology): “This model is based on Zeiler and Fergus’s fast model [25] but with the SPP layer [11].”
A Theory of Generative ConvNet (2016, methodology): “…the filters in the bottomup encoding become the basis functions in the top-down decoding, and the binary activation variables detected by the filters in the bottom-up convolution process become the coefficients of the basis functions in the top-down deconvolution process (Zeiler & Fergus, 2014).”
Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation (2014, background): “The best performing algorithms today for many vision tasks in general [38, 28], and human pose estimation in particular ([36, 16, 35]) are based on deep convolutional networks.”
DenseCap: Fully Convolutional Localization Networks for Dense Captioning (2015, background): “In the last few years there has been significant progress in image classification [38, 26, 52, 44], where the task is to assign one label to an image.”
Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation (2014, methodology): “We use the 15 layer convolutional architecture of [8], trained on the ImageNet 2012 dataset [9].”
BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation (2015, background): “With the continuously improving deep convolutional models [31, 25, 3, 13, 26, 27, 14], the accuracy of these vision tasks also improves thanks to the more representative generic features learned from large-scale datasets.”