Image segmentation approach for realizing zoomable streaming HEVC video

Yuri Umezaki, Satoshi Goto

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

With the popularity of mobile devices such as mobile phones and cameras, zooming and displaying an arbitrary region of interest (ROI) from a high-resolution video has gained considerable importance. In this study, we implement two methods - partial decoding and tiled encoding - for ROI-based streaming in High Efficiency Video Coding (HEVC). In partial decoding, only the ROI and dependence area are decoded; this approach reduces the calculation cost. In tiled encoding, video frames are split into a grid of tiles, each of which is encoded independently; this approach requires a compatible decoder, and it reduces the streaming bitrate and decoding calculation cost. We evaluated these two methods in terms of the decoding calculation cost, bandwidth efficiency, and video quality. Experimental results show that partial decoding reduced the decoding calculation cost by 35-55%. Tiled encoding showed optimal decoding calculation cost and bandwidth efficiency for a tile size of 128 pixels.

Original languageEnglish
Title of host publicationICICS 2013 - Conference Guide of the 9th International Conference on Information, Communications and Signal Processing
PublisherIEEE Computer Society
DOIs
Publication statusPublished - 2013
Event9th International Conference on Information, Communications and Signal Processing, ICICS 2013 - Tainan
Duration: 2013 Dec 102013 Dec 13

Other

Other9th International Conference on Information, Communications and Signal Processing, ICICS 2013
CityTainan
Period13/12/1013/12/13

Fingerprint

Image coding
Image segmentation
Decoding
Costs
Tile
Bandwidth
Mobile phones
Mobile devices
Pixels
Cameras

Keywords

  • Cropping
  • HEVC
  • Mobile streaming
  • Region of interest
  • Zoomable video

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing

Cite this

Umezaki, Y., & Goto, S. (2013). Image segmentation approach for realizing zoomable streaming HEVC video. In ICICS 2013 - Conference Guide of the 9th International Conference on Information, Communications and Signal Processing [6782830] IEEE Computer Society. https://doi.org/10.1109/ICICS.2013.6782830

Image segmentation approach for realizing zoomable streaming HEVC video. / Umezaki, Yuri; Goto, Satoshi.

ICICS 2013 - Conference Guide of the 9th International Conference on Information, Communications and Signal Processing. IEEE Computer Society, 2013. 6782830.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Umezaki, Y & Goto, S 2013, Image segmentation approach for realizing zoomable streaming HEVC video. in ICICS 2013 - Conference Guide of the 9th International Conference on Information, Communications and Signal Processing., 6782830, IEEE Computer Society, 9th International Conference on Information, Communications and Signal Processing, ICICS 2013, Tainan, 13/12/10. https://doi.org/10.1109/ICICS.2013.6782830
Umezaki Y, Goto S. Image segmentation approach for realizing zoomable streaming HEVC video. In ICICS 2013 - Conference Guide of the 9th International Conference on Information, Communications and Signal Processing. IEEE Computer Society. 2013. 6782830 https://doi.org/10.1109/ICICS.2013.6782830
Umezaki, Yuri ; Goto, Satoshi. / Image segmentation approach for realizing zoomable streaming HEVC video. ICICS 2013 - Conference Guide of the 9th International Conference on Information, Communications and Signal Processing. IEEE Computer Society, 2013.
@inproceedings{74d36bff26bb42c882fd18f715b6dc1d,
title = "Image segmentation approach for realizing zoomable streaming HEVC video",
abstract = "With the popularity of mobile devices such as mobile phones and cameras, zooming and displaying an arbitrary region of interest (ROI) from a high-resolution video has gained considerable importance. In this study, we implement two methods - partial decoding and tiled encoding - for ROI-based streaming in High Efficiency Video Coding (HEVC). In partial decoding, only the ROI and dependence area are decoded; this approach reduces the calculation cost. In tiled encoding, video frames are split into a grid of tiles, each of which is encoded independently; this approach requires a compatible decoder, and it reduces the streaming bitrate and decoding calculation cost. We evaluated these two methods in terms of the decoding calculation cost, bandwidth efficiency, and video quality. Experimental results show that partial decoding reduced the decoding calculation cost by 35-55{\%}. Tiled encoding showed optimal decoding calculation cost and bandwidth efficiency for a tile size of 128 pixels.",
keywords = "Cropping, HEVC, Mobile streaming, Region of interest, Zoomable video",
author = "Yuri Umezaki and Satoshi Goto",
year = "2013",
doi = "10.1109/ICICS.2013.6782830",
language = "English",
booktitle = "ICICS 2013 - Conference Guide of the 9th International Conference on Information, Communications and Signal Processing",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Image segmentation approach for realizing zoomable streaming HEVC video

AU - Umezaki, Yuri

AU - Goto, Satoshi

PY - 2013

Y1 - 2013

N2 - With the popularity of mobile devices such as mobile phones and cameras, zooming and displaying an arbitrary region of interest (ROI) from a high-resolution video has gained considerable importance. In this study, we implement two methods - partial decoding and tiled encoding - for ROI-based streaming in High Efficiency Video Coding (HEVC). In partial decoding, only the ROI and dependence area are decoded; this approach reduces the calculation cost. In tiled encoding, video frames are split into a grid of tiles, each of which is encoded independently; this approach requires a compatible decoder, and it reduces the streaming bitrate and decoding calculation cost. We evaluated these two methods in terms of the decoding calculation cost, bandwidth efficiency, and video quality. Experimental results show that partial decoding reduced the decoding calculation cost by 35-55%. Tiled encoding showed optimal decoding calculation cost and bandwidth efficiency for a tile size of 128 pixels.

AB - With the popularity of mobile devices such as mobile phones and cameras, zooming and displaying an arbitrary region of interest (ROI) from a high-resolution video has gained considerable importance. In this study, we implement two methods - partial decoding and tiled encoding - for ROI-based streaming in High Efficiency Video Coding (HEVC). In partial decoding, only the ROI and dependence area are decoded; this approach reduces the calculation cost. In tiled encoding, video frames are split into a grid of tiles, each of which is encoded independently; this approach requires a compatible decoder, and it reduces the streaming bitrate and decoding calculation cost. We evaluated these two methods in terms of the decoding calculation cost, bandwidth efficiency, and video quality. Experimental results show that partial decoding reduced the decoding calculation cost by 35-55%. Tiled encoding showed optimal decoding calculation cost and bandwidth efficiency for a tile size of 128 pixels.

KW - Cropping

KW - HEVC

KW - Mobile streaming

KW - Region of interest

KW - Zoomable video

UR - http://www.scopus.com/inward/record.url?scp=84899107232&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899107232&partnerID=8YFLogxK

U2 - 10.1109/ICICS.2013.6782830

DO - 10.1109/ICICS.2013.6782830

M3 - Conference contribution

BT - ICICS 2013 - Conference Guide of the 9th International Conference on Information, Communications and Signal Processing

PB - IEEE Computer Society

ER -