GPU-based real-time super-resolution system for high-quality UHD video up-conversion

Lee, Dae Yeol; Lee, Jooyoung; Choi, Ji Hoon; Kim, Jong Ok; Kim, Hui Yong; Choi, Jin Soo

doi:10.1007/s11227-017-2136-1

Cited by 5 publications

(2 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The autoregressive approaches [15,17] estimate the distribution of a current latent representation using its adjacent known representations, thus leading to improve the compression performance by removing the correlations between the current latent representation and its neighbors. Although their methods effectively remove the spatial and inter-channel correlations among the transformed representations, our global context exploitation further improves the coding efficiency by removing the remaining spatial correlations across a wider area of each input image, motivated by the known wisdom [8,14] that self-similarity exists in the natural images.…”

Section: Introductionmentioning

confidence: 99%

An End-to-End Joint Learning Scheme of Image Compression and Quality Enhancement with Improved Entropy Minimization

Lee¹,

Cho²,

Kim³

2019

Preprint

Self Cite

View full text Add to dashboard Cite

Recently, learned image compression methods have been actively studied. Among them, entropy-minimization based approaches have achieved superior results compared to conventional image codecs such as BPG and JPEG2000. However, the quality enhancement and rate-minimization are conflictively coupled in the process of image compression. That is, maintaining high image quality entails less compression and vice versa. However, by jointly training separate quality enhancement in conjunction with image compression, the coding efficiency can be improved. In this paper, we propose a novel hybrid architecture of jointly learning conflictively coupled image compression and quality enhancement, called JointIQ-Net, as well as entropy model improvement, thus achieving significantly improved coding efficiency against the previous methods. Our proposed JointIQ-Net combines an image compression sub-net and an image quality enhancement sub-net in a cascade, both of which are end-to-end trained in a combined manner within the JointIQ-Net. Also the JointIQ-Net benefits from improved entropy-minimization that newly adopts a Gussian Mixture Model (GMM) and further exploits global context to estimate the probabilities of latent representations. In order to show the effectiveness of our proposed JointIQ-Net, extensive experiments have been performed, and showed that the JointIQ-Net achieves a remarkable performance improvement in coding efficiency in terms of both PSNR and MS-SSIM, compared to the previous learned image compression methods and the conventional codecs such as VTM 7.1 (intra), BPG, and JPEG2000. To the best of our knowledge, this is the first end-to-end optimized image compression method that outperforms VTM 7.1 (intra), the latest reference software of the VVC standard, in terms of the PSNR and MS-SSIM.

show abstract

Section: Introductionmentioning

confidence: 99%

An End-to-End Joint Learning Scheme of Image Compression and Quality Enhancement with Improved Entropy Minimization

Lee¹,

Cho²,

Kim³

2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recently, researchers pay more attention to SR methods based on neural network (NC) [11,12], which can be trained from data to approximate complex nonlinear functions, while these algorithms require significant computational resources to achieve real-time throughput. Graphic processing unit (GPU) is used to accelerate, as done by Hu et al [13][14][15][16].…”

Section: Introductionmentioning

confidence: 99%

Online multi-frame super-resolution of image sequences

Liang

Liu

et al. 2018

J Image Video Proc.

View full text Add to dashboard Cite

Multi-frame super-resolution recovers a high-resolution (HR) image from a sequence of low-resolution (LR) images. In this paper, we propose an algorithm that performs multi-frame super-resolution in an online fashion. This algorithm processes only one low-resolution image at a time instead of co-processing all LR images which is adopted by stateof-the-art super-resolution techniques. Our algorithm is very fast and memory efficient, and simple to implement. In addition, we employ a noise-adaptive parameter in the classical steepest gradient optimization method to avoid noise amplification and overfitting LR images. Experiments with simulated and real-image sequences yield promising results.

show abstract

Face video Super Resolution using Deep Convolutional Neural Network

Deshmukh

Rani

2019

2019 5th International Conference on Computing, Communication, Control and Automation (ICCUBEA)

View full text Add to dashboard Cite

GPU-based real-time super-resolution system for high-quality UHD video up-conversion

Cited by 5 publications

References 23 publications

An End-to-End Joint Learning Scheme of Image Compression and Quality Enhancement with Improved Entropy Minimization

An End-to-End Joint Learning Scheme of Image Compression and Quality Enhancement with Improved Entropy Minimization

Online multi-frame super-resolution of image sequences

Face video Super Resolution using Deep Convolutional Neural Network

Contact Info

Product

Resources

About