A Powerful Generative Model Using Random Weights for the Deep Image Representation

By Kun He, Yan Wang, John E. Hopcroft. 2016

79 citations recorded by the source corpus.

To what extent is the success of deep visualization due to the training? Could we do deep visualization using untrained, random weight networks? To address this issue, we explore new and powerful generative models for three popular deep visualization tasks using untrained, random weight convolutional neural networks. First we invert representations in feature spaces and reconstruct images from white noise inputs. The reconstruction quality is statistically higher than that of the same method applied on well trained networks with the same architecture. Next we synthesize textures using scaled correlations of representations in multiple layers and our results are almost indistinguishable with the original natural texture and the synthesized textures based on the trained network. Third, by recasting the content of an image in the style of various artworks, we create artistic images with high perceptual quality, highly competitive to the prior work of Gatys et al. on pretrained networks. To our knowledge this is the first demonstration of image representations using untrained deep neural networks. Our work provides a new and fascinating tool to study the representation of deep network architecture and sheds light on new understandings on deep visualization.

Subfields: Cognitive Neuroscience, 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 Cognitive Neuroscience

In Computer Vision and Pattern Recognition

Citations in the corpus

Weight Agnostic Neural Networks (2019, background): “Recent work [45, 117] demonstrated that even randomly-initialized CNNs can be used effectively for image processing tasks such as superresolution, inpainting and style transfer.”
On Using Backpropagation for Speech Texture Generation and Voice Conversion (2017, background): “In contrast, image texture synthesis and style transfer have been reported to work with randomly initialized networks [30].”
The History Began from AlexNet: A Comprehensive Survey on Deep Learning Approaches (2018, unknown): “Due to their composition, many layers of DNNs are more capable at representing highly varying nonlinear functions compared to shallow learning approaches [56, 57, and 58].”
Transformation-Grounded Image Generation Network for Novel 3D View Synthesis (2017, methodology): “We tried both a pre-trained loss network and a network with random weights as suggested in [15, 39].”
Adversarial Reprogramming of Neural Networks (2018, background): “Other work has demonstrated that it is possible to use random networks as generative models for images (Ustyuzhaninov et al., 2016; He et al., 2016), further supporting their potential richness.”
Convolutional Neural Networks (2017, unknown): “Also, He et al. (2016) empirically showed that random weight CNNs can do image reconstruction well.”