Corpus ID: 1450294

Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps

@article{Simonyan2014DeepIC,
  title={Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps},
  author={Karen Simonyan and Andrea Vedaldi and Andrew Zisserman},
  journal={CoRR},
  year={2014},
  volume={abs/1312.6034}
}
This paper addresses the visualisation of image classification models, learnt using deep Convolutional Networks (ConvNets. [...] Key Method The first one generates an image, which maximises the class score [Erhan et al., 2009], thus visualising the notion of the class, captured by a ConvNet. The second technique computes a class saliency map, specific to a given image and class. We show that such maps can be employed for weakly supervised object segmentation using classification ConvNets. Finally, we establish…Expand
MACE: Model Agnostic Concept Extractor for Explaining Image Classification Networks
TLDR
This work proposes MACE: a Model Agnostic Concept Extractor, which can explain the working of a convolutional network through smaller concepts, and demonstrates that the concepts extracted increase the human interpretability of the explanations, and are faithful to the underlying pre-trained black-box model. Expand
Visualizing and analyzing convolution neural networks with gradient information
TLDR
The gradient-based visualization method is expanded by emphasizing the ability of CNNs in extracting local features, visualizes and analyzes the internal behavior of trained CNNs based on computing the gradient of the class score with respect to every layer ofCNNs. Expand
Learning FRAME Models Using CNN Filters for Knowledge Visualization
TLDR
This paper proposes to learn the generative FRAME (Filters, Random field, And Maximum Entropy) model using the highly expressive filters pre-learned by the CNN at the convolutional layers, and explains how this model corresponds to a CNN unit at a layer above the layer of filters employed by the model. Expand
Segmentation and Shape Extraction from Convolutional Neural Networks
TLDR
A novel method for creating high-resolution class activation maps from a given deep convolutional neural network which was trained for image classification that is competitive with state-of-the-art weakly supervised segmentation methods on datasets on which the latter have been trained. Expand
Deep Learning for Object Saliency Detection and Image Segmentation
TLDR
This work uses a gradient descent method to iteratively modify an input image based on the pixel-wise gradients to reduce a cost function measuring the class-specific objectness of the image, and uses the computed saliency maps for image segmentation. Expand
Smooth Grad-CAM++: An Enhanced Inference Level Visualization Technique for Deep Convolutional Neural Network Models
TLDR
The Smooth Grad-CAM++ technique provides the capability of either visualizing a layer, subset of feature maps, or subset of neurons within a feature map at each instance at the inference level (model prediction process). Expand
Visualization Methods for Image Transformation Convolutional Neural Networks
TLDR
This paper uses the knowledge obtained from the visualization of an image restoration CNN to improve the architecture’s efficiency with no significant degradation of its performance. Expand
Visualizing Deep Convolutional Neural Networks Using Natural Pre-images
TLDR
This paper studies several landmark representations, both shallow and deep, by a number of complementary visualization techniques based on the concept of “natural pre-image”, and shows that several layers in CNNs retain photographically accurate information about the image, with different degrees of geometric and photometric invariance. Expand
Visualizing and Understanding Convolutional Networks
TLDR
A novel visualization technique is introduced that gives insight into the function of intermediate feature layers and the operation of the classifier in large Convolutional Network models, used in a diagnostic role to find model architectures that outperform Krizhevsky et al on the ImageNet classification benchmark. Expand
Saliency-Driven Class Impressions For Feature Visualization Of Deep Neural Networks
TLDR
A saliency-driven approach to visualize discriminative features that are considered most important for a given task, and restricts the algorithm to develop a single object per image, which helps further in extracting features of high confidence, and also results in better visualizations. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 17 REFERENCES
Visualizing and Understanding Convolutional Networks
TLDR
A novel visualization technique is introduced that gives insight into the function of intermediate feature layers and the operation of the classifier in large Convolutional Network models, used in a diagnostic role to find model architectures that outperform Krizhevsky et al on the ImageNet classification benchmark. Expand
ImageNet classification with deep convolutional neural networks
TLDR
A large, deep convolutional neural network was trained to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes and employed a recently developed regularization method called "dropout" that proved to be very effective. Expand
Visualizing Higher-Layer Features of a Deep Network
TLDR
This paper contrast and compare several techniques applied on Stacked Denoising Autoencoders and Deep Belief Networks, trained on several vision datasets, and shows that good qualitative interpretations of high level features represented by such models are possible at the unit level. Expand
Improving the Fisher Kernel for Large-Scale Image Classification
TLDR
In an evaluation involving hundreds of thousands of training images, it is shown that classifiers learned on Flickr groups perform surprisingly well and that they can complement classifier learned on more carefully annotated datasets. Expand
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. Expand
Building high-level features using large scale unsupervised learning
TLDR
Contrary to what appears to be a widely-held intuition, the experimental results reveal that it is possible to train a face detector without having to label images as containing a face or not. Expand
Large scale visual recognition
TLDR
A novel learning technique is proposed that scales logarithmically with the number of classes in both training and testing, improving both accuracy and efficiency of the previous state of the art while reducing training time by 31 fold on 10 thousand classes. Expand
A discriminatively trained, multiscale, deformable part model
TLDR
A discriminatively trained, multiscale, deformable part model for object detection, which achieves a two-fold improvement in average precision over the best performance in the 2006 PASCAL person detection challenge and outperforms the best results in the 2007 challenge in ten out of twenty categories. Expand
A Fast Learning Algorithm for Deep Belief Nets
TLDR
A fast, greedy algorithm is derived that can learn deep, directed belief networks one layer at a time, provided the top two layers form an undirected associative memory. Expand
Gradient-based learning applied to document recognition
TLDR
This paper reviews various methods applied to handwritten character recognition and compares them on a standard handwritten digit recognition task, and Convolutional neural networks are shown to outperform all other techniques. Expand
...
1
2
...