#### Introduction
* The paper presents gradient computation based techniques to visualise image classification models.
* [Link to the paper](https://arxiv.org/abs/1312.6034)
#### Experimental Setup
* Single deep convNet trained on ILSVRC-2013 dataset (1.2M training images and 1000 classes).
* Weight layer configuration is: conv64-conv256-conv256-conv256-conv256-full4096-full4096-full1000.
#### Class Model Visualisation
* Given a learnt ConvNet and a class (of interest), start with the zero image and perform optimisation by back propagating with respect to the input image (keeping the ConvNet weights constant).
* Add the mean image (for training set) to the resulting image.
* The paper used unnormalised class scores so that optimisation focuses on increasing the score of target class and not decreasing the score of other classes.
#### Image-Specific Class Saliency Visualisation
* Given an image, class of interest, and trained ConvNet, rank the pixels of the input image based on their influence on class scores.
* Derivative of the class score with respect to image gives an estimate of the importance of different pixels for the class.
* The magnitude of derivative also indicated how much each pixel needs to be changed to improve the class score.
##### Class Saliency Extraction
* Find the derivative of the class score with respect with respect to the input image.
* This would result in one single saliency map per colour channel.
* To obtain a single saliency map, take the maximum magnitude of derivative across all colour channels.
##### Weakly Supervised Object Localisation
* The saliency map for an image provides a rough encoding of the location of the object of the class of interest.
* Given an image and its saliency map, an object segmentation map can be computed using GraphCut colour segmentation.
* Color continuity cues are needed as saliency maps might capture only the most dominant part of the object in the image.
* This weakly supervised approach achieves 46.4% top-5 error on the test set of ILSVRC-2013.
#### Relation to Deconvolutional Networks
* DeconvNet-based reconstruction of the $n^{th}$ layer input is similar to computing the gradient of the visualised neuron activity $f$ with respect to the input layer.
* One difference is in the way RELU neurons are treated:
* In DeconvNet, the sign indicator (for the derivative of RELU) is computed on output reconstruction while in this paper, the sign indicator is computed on the layer input.
This paper presents methods for visualizing the behaviour of an object recognition convolutional neural network. The first method generates a "canonical image" for a given class that the network can recognize. The second generates a saliency map for a given input image and specified class, that illustrates the part of the image (pixels) that influence the most the given class's output probability. This can be used to seed a graphcut segmentation and localize objects of that class in the input image. Finally, a connection between the saliency map method and the work of Zeiler and Fergus on using deconvolutions to visualize deep networks is established.
This paper attempts to understand the representations learnt by deep
convolutional neural networks by introducing two interpretable visualization
techniques. Main contributions:
- Class model visualizations
- These are obtained by making numerical optimizations in the input
space to maximize the class score. Gradients are calculated wrt input
and are used to update the input image (initialized with zero image),
while weights are kept fixed to those obtained from training.
- Image-specific saliency map visualizations
- These are approximated by using the same gradient as before (gradient
of class score wrt input). The absolute pixel-wise max across channels produces
the saliency map.
- Relation between DeconvNet and optimization-based visualizations
- Visualizations using DeconvNet are the same as gradient-based methods except
for ReLU. In regular backprop, gradients flow through ReLU to units with positive
input activations, whereas in case of a DeconvNet, it is computed on positive output
reconstructions.
## Strengths
- The visualization techniques are simple ideas and the results are interpretable. They show
that the method proposed by Erhan et al. in an unsupervised setting is useful to CNNs trained
in a supervised manner as well.
- The image-specific class saliency can be interpreted as those pixels which need to be changed
the least to have a maximum impact on the classification score.
- The relation between DeconvNet visualizations and optimization-based visualizations is
insightful.
## Weaknesses / Notes
- The thinking behind initializing with zero image and L2 regularization in class model
visualizations was missing.