Hypernetwork functional image representation

Abstract

Motivated by the human way of memorizing images we introduce their functional representation, where an image is represented by a neural network. For this purpose, we construct a hypernetwork which takes an image and returns weights to the target network, which maps point from the plane (representing positions of the pixel) into its corresponding color in the image. Since the obtained representation is continuous, one can easily inspect the image at various resolutions and perform on it arbitrary continuous operations. Moreover, by inspecting interpolations we show that such representation has some properties characteristic to generative models. To evaluate the proposed mechanism experimentally, we apply it to image super-resolution problem. Despite using a single model for various scaling factors, we obtained results comparable to existing super-resolution methods.

References in corpus (13)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Auto-Encoding Variational Bayes
• Neural Architecture Search with Reinforcement Learning
• Beyond Correlation Filters: Learning Continuous Convolution Operators for Visual Tracking
• Unsupervised Image-to-Image Translation Networks
• SMASH: One-Shot Model Architecture Search through HyperNetworks
• Multiplicative Normalizing Flows for Variational Bayesian Neural Networks
• Wasserstein Auto-Encoders
• Graph HyperNetworks for Neural Architecture Search
• HyperNetworks
• Stochastic Hyperparameter Optimization through Hypernetworks
• Evolutionary-Neural Hybrid Agents for Architecture Search
• Stochastic Maximum Likelihood Optimization via Hypernetworks

Cited by in corpus (14)

• A Brief Review of Hypernetworks in Deep Learning
• Implicit Full Waveform Inversion with Deep Neural Representation
• A newcomer's guide to deep learning for inverse design in nano-photonics
• Learning the Effect of Registration Hyperparameters with HyperMorph
• UltraSR: Spatial Encoding is a Missing Key for Implicit Image Function-based Arbitrary-Scale Super-Resolution
• Principled Weight Initialization for Hypernetworks
• Learning the Pareto Front with Hypernetworks
• Hyper-Convolution Networks for Biomedical Image Segmentation
• Leveraging Hypernetworks and Learnable Kernels for Consumer Energy Forecasting Across Diverse Consumer Types
• Regularization-Agnostic Compressed Sensing MRI Reconstruction with Hypernetworks
• On Infinite-Width Hypernetworks
• HyperFlow: Representing 3D Objects as Surfaces
• Block Contextual MDPs for Continual Learning
• Hypernetwork approach to generating point clouds