Randomness in Deconvolutional Networks for Visual Representation

TL;DR

This paper investigates the properties of untrained CNNs using deconvolutional networks and randomization, revealing that random features contain rich information and that the architecture allows effective image reconstruction.

Contribution

It introduces a novel analysis of untrained CNNs with DCNs, showing their potential for image reconstruction and providing both empirical and theoretical insights.

Findings

Random CNN features encode rich information.

Untrained CNNs can be inverted to reconstruct images.

The architecture facilitates effective image reconstruction even without training.

Abstract

Toward a deeper understanding on the inner work of deep neural networks, we investigate CNN (convolutional neural network) using DCN (deconvolutional network) and randomization technique, and gain new insights for the intrinsic property of this network architecture. For the random representations of an untrained CNN, we train the corresponding DCN to reconstruct the input images. Compared with the image inversion on pre-trained CNN, our training converges faster and the yielding network exhibits higher quality for image reconstruction. It indicates there is rich information encoded in the random features; the pre-trained CNN may discard information irrelevant for classification and encode relevant features in a way favorable for classification but harder for reconstruction. We further explore the property of the overall random CNN-DCN architecture. Surprisingly, images can be inverted with satisfactory quality. Extensive empirical evidence as well as theoretical analysis are provided.