Please use this identifier to cite or link to this item:
Title: Unsupervised real image super-resolution via Generative Variational AutoEncoder
Author(s): Siu, Wan Chi 
Liu, Zhisong 
Author(s): Wang, L.-W.
Li, C.-T.
Cani, M.-P.
Chan, Y.-L.
Issue Date: 2020
Publisher: IEEE
Related Publication(s): Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops
Benefited from the deep learning, image Super-Resolution has been one of the most developing research fields in computer vision. Depending upon whether using a discriminator or not, a deep convolutional neural network can provide an image with high fidelity or better perceptual quality. Due to the lack of ground truth images in real life, people prefer a photo-realistic image with low fidelity to a blurry image with high fidelity. In this paper, we revisit the classic example based image super-resolution approaches and come up with a novel generative model for perceptual image super-resolution. Given that real images contain various noise and artifacts, we propose a joint image denoising and super-resolution model via Variational AutoEncoder. We come up with a conditional variational autoencoder to encode the reference for dense feature vector which can then be transferred to the decoder for target image denoising. With the aid of the discriminator, an additional overhead of super-resolution subnetwork is attached to super-resolve the denoised image with photo-realistic visual quality. We participated the NTIRE2020 Real Image Super-Resolution Challenge. Experimental results show that by using the proposed approach, we can obtain enlarged images with clean and pleasant features compared to other supervised methods. We also compared our approach with state-of-the-art methods on various datasets to demonstrate the efficiency of our proposed unsupervised super-resolution model.
CIHE Affiliated Publication: No
Appears in Collections:CIS Publication

Files in This Item:
File Description SizeFormat
View Online212 BHTMLView/Open
SFX Query Show full item record

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.