Video Compression based on Spatio-Temporal Resolution Adaptation

Mariana Afonso; Aaron Zhang; David Bull

doi:10.1109/TCSVT.2018.2878952

Video Compression based on Spatio-Temporal Resolution Adaptation

Mariana Afonso, Aaron Zhang, David Bull

Research output: Contribution to journal › Article (Academic Journal) › peer-review

65 Citations (Scopus)

1064 Downloads (Pure)

Abstract

A Video Compression framework based on Spatio-Temporal Resolution Adaptation (ViSTRA) is proposed, which dynamically resamples the input video spatially and temporally during encoding, based on a quantisation-resolution decision, and reconstructs the full resolution video at the decoder. Temporal upsampling is performed using frame repetition, whereas a Convolutional Neural Network (CNN) superresolution model is employed for spatial resolution upsampling. ViSTRA has been integrated into the High Efficiency Video Coding (HEVC) reference software (HM 16.14). Experimental results verified via an international challenge show significant improvements, with BD-rate gains of 15% based on PSNR and an average MOS difference of 0.5 based on subjective visual quality tests.

Original language	English
Pages (from-to)	275-280
Number of pages	6
Journal	IEEE Transactions on Circuits and Systems for Video Technology
Volume	29
Issue number	1
Early online date	31 Oct 2018
DOIs	https://doi.org/10.1109/TCSVT.2018.2878952
Publication status	Published - 1 Jan 2019

Keywords

Video compression
spatial resolution adaptation
temporal resolution adaptation
perceptual video compression
CNN-based super-resolution

Access to Document

10.1109/TCSVT.2018.2878952

Full-text PDF (final published version)
This is the final published version of the article (version of record). It first appeared online via IEEE at https://ieeexplore.ieee.org/document/8517114. Please refer to any applicable terms of use of the publisher.
Final published version, 2.15 MBLicence: CC BY

Handle.net

Persistent link

Embargoed Document

Full-text PDF (accepted author manuscript)
Accepted author manuscript, 953 KB

65 Citations
1 Conference Contribution (Conference Proceeding)

Compressing Deep Image Super-Resolution Models
Jiang, Y., Nawala, J. T., Zhang, F. & Bull, D. R., 26 Jun 2024, 2024 Picture Coding Symposium (PCS). Institute of Electrical and Electronics Engineers (IEEE), 5 p. (Picture Coding Symposium, PCS).
Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

Open Access
File
2 Citations (Scopus)

34 Downloads (Pure)

1 Finished

Vision for the Future-Full
Bull, D. R. (Principal Investigator)
1/02/15 → 31/01/20
Project: Research

Intelligent Resampling Methods for Video Compression
Fernandez Afonso, M. (Author), Agrafiotis, D. (Supervisor) & Bull, D. (Supervisor), 25 Jun 2019
Student thesis: Doctoral Thesis › Doctor of Philosophy (PhD)

File

Professor David R Bull
- Dave.Bullbristol.acuk
- School of Computer Science - Professor of Signal Processing
- Visual Information Laboratory
- Bristol Vision Institute
Person: Academic , Group lead

Cite this

@article{64c288ab90a0477f84c617e3ffc8c06f,

title = "Video Compression based on Spatio-Temporal Resolution Adaptation",

abstract = "A Video Compression framework based on Spatio-Temporal Resolution Adaptation (ViSTRA) is proposed, which dynamically resamples the input video spatially and temporally during encoding, based on a quantisation-resolution decision, and reconstructs the full resolution video at the decoder. Temporal upsampling is performed using frame repetition, whereas a Convolutional Neural Network (CNN) superresolution model is employed for spatial resolution upsampling. ViSTRA has been integrated into the High Efficiency Video Coding (HEVC) reference software (HM 16.14). Experimental results verified via an international challenge show significant improvements, with BD-rate gains of 15% based on PSNR and an average MOS difference of 0.5 based on subjective visual quality tests.",

keywords = "Video compression, spatial resolution adaptation, temporal resolution adaptation, perceptual video compression, CNN-based super-resolution",

author = "Mariana Afonso and Aaron Zhang and David Bull",

year = "2019",

month = jan,

day = "1",

doi = "10.1109/TCSVT.2018.2878952",

language = "English",

volume = "29",

pages = "275--280",

journal = "IEEE Transactions on Circuits and Systems for Video Technology",

issn = "1051-8215",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

number = "1",

}

TY - JOUR

T1 - Video Compression based on Spatio-Temporal Resolution Adaptation

AU - Afonso, Mariana

AU - Zhang, Aaron

AU - Bull, David

PY - 2019/1/1

Y1 - 2019/1/1

N2 - A Video Compression framework based on Spatio-Temporal Resolution Adaptation (ViSTRA) is proposed, which dynamically resamples the input video spatially and temporally during encoding, based on a quantisation-resolution decision, and reconstructs the full resolution video at the decoder. Temporal upsampling is performed using frame repetition, whereas a Convolutional Neural Network (CNN) superresolution model is employed for spatial resolution upsampling. ViSTRA has been integrated into the High Efficiency Video Coding (HEVC) reference software (HM 16.14). Experimental results verified via an international challenge show significant improvements, with BD-rate gains of 15% based on PSNR and an average MOS difference of 0.5 based on subjective visual quality tests.

AB - A Video Compression framework based on Spatio-Temporal Resolution Adaptation (ViSTRA) is proposed, which dynamically resamples the input video spatially and temporally during encoding, based on a quantisation-resolution decision, and reconstructs the full resolution video at the decoder. Temporal upsampling is performed using frame repetition, whereas a Convolutional Neural Network (CNN) superresolution model is employed for spatial resolution upsampling. ViSTRA has been integrated into the High Efficiency Video Coding (HEVC) reference software (HM 16.14). Experimental results verified via an international challenge show significant improvements, with BD-rate gains of 15% based on PSNR and an average MOS difference of 0.5 based on subjective visual quality tests.

KW - Video compression

KW - spatial resolution adaptation

KW - temporal resolution adaptation

KW - perceptual video compression

KW - CNN-based super-resolution

U2 - 10.1109/TCSVT.2018.2878952

DO - 10.1109/TCSVT.2018.2878952

M3 - Article (Academic Journal)

SN - 1051-8215

VL - 29

SP - 275

EP - 280

JO - IEEE Transactions on Circuits and Systems for Video Technology

JF - IEEE Transactions on Circuits and Systems for Video Technology

IS - 1

ER -

Video Compression based on Spatio-Temporal Resolution Adaptation

Abstract

Keywords

Access to Document

Handle.net

Embargoed Document

Fingerprint

Research output

Compressing Deep Image Super-Resolution Models

Projects

Vision for the Future-Full

Student theses

Intelligent Resampling Methods for Video Compression

Profiles

Professor David R Bull

Cite this