Guided Deep Decoder: Unsupervised Image Pair Fusion
Published in
European Conference on Computer Vision(2020)External Links:
TL
TL;DR
A guided deep decoder network is proposed as a general prior that can achieve state-of-the-art performance in various image fusion problems and allows the network parameters to be optimized in an unsupervised way without training data.
Abstract
The fusion of input and guidance images that have a tradeoff in their information (e.g., hyperspectral and RGB image fusion or pansharpening) can be interpreted as one general problem. However, previous studies applied a task-specific handcrafted prior and did not address the problems with a unified approach. To address this limitation, in this study, we propose a guided deep decoder network as a general prior. The proposed network is composed of an encoder-decoder network that exploits multi-scale features of a guidance image and a deep decoder network that generates an output image. The two networks are connected by feature refinement units to embed the multi-scale features of the guidance image into the deep decoder network. The proposed network allows the network parameters to be optimized in an unsupervised way without training data. Our results show that the proposed network can achieve state-of-the-art performance in various image fusion problems.
References49 items
1
The Spectral Bias of the Deep Image Prior
2
Learning to Predict Layout-to-image Conditional Convolutions for Semantic Image Synthesis
3
Dynamic PET Image Reconstruction Using Nonnegative Matrix Factorization Incorporated With Deep Image Prior
4
Hyperspectral Image Super-Resolution With Optimized RGB Guidance
5