Model-based referenceless quality metric of 3D synthesized images using local image description
New challenges have been brought out along with the emerging of 3D-related technologies, such as virtual reality, augmented reality (AR), and mixed reality. Free viewpoint video (FVV), due to its applications in remote surveillance, remote education, and so on, based on the flexible selection of dir...
Saved in:
Main Authors: | , , , , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
2020
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/142320 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-142320 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1423202020-06-19T03:36:20Z Model-based referenceless quality metric of 3D synthesized images using local image description Gu, Ke Jakhetiya, Vinit Qiao, Jun-Fei Li, Xiaoli Lin, Weisi Thalmann, Daniel School of Computer Science and Engineering BeingThere Centre - Institute for Media Innovation Engineering::Computer science and engineering Quality Assessment No-reference New challenges have been brought out along with the emerging of 3D-related technologies, such as virtual reality, augmented reality (AR), and mixed reality. Free viewpoint video (FVV), due to its applications in remote surveillance, remote education, and so on, based on the flexible selection of direction and viewpoint, has been perceived as the development direction of next-generation video technologies and has drawn a wide range of researchers' attention. Since FVV images are synthesized via a depth image-based rendering (DIBR) procedure in the "blind" environment (without reference images), a reliable real-time blind quality evaluation and monitoring system is urgently required. But existing assessment metrics do not render human judgments faithfully mainly because geometric distortions are generated by DIBR. To this end, this paper proposes a novel referenceless quality metric of DIBR-synthesized images using the autoregression (AR)-based local image description. It was found that, after the AR prediction, the reconstructed error between a DIBR-synthesized image and its AR-predicted image can accurately capture the geometry distortion. The visual saliency is then leveraged to modify the proposed blind quality metric to a sizable margin. Experiments validate the superiority of our no-reference quality method as compared with prevailing full-, reduced-, and no-reference models. MOE (Min. of Education, S’pore) 2020-06-19T03:36:20Z 2020-06-19T03:36:20Z 2017 Journal Article Gu, K., Jakhetiya, V., Qiao, J.-F., Li, X., Lin, W., & Thalmann, D. (2018). Model-based referenceless quality metric of 3D synthesized images using local image description. IEEE Transactions on Image Processing, 27(1), 394-405. doi:10.1109/TIP.2017.2733164 1057-7149 https://hdl.handle.net/10356/142320 10.1109/TIP.2017.2733164 28767368 2-s2.0-85028915580 1 27 394 405 en IEEE Transactions on Image Processing © 2017 IEEE. All rights reserved. |
institution |
Nanyang Technological University |
building |
NTU Library |
country |
Singapore |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Computer science and engineering Quality Assessment No-reference |
spellingShingle |
Engineering::Computer science and engineering Quality Assessment No-reference Gu, Ke Jakhetiya, Vinit Qiao, Jun-Fei Li, Xiaoli Lin, Weisi Thalmann, Daniel Model-based referenceless quality metric of 3D synthesized images using local image description |
description |
New challenges have been brought out along with the emerging of 3D-related technologies, such as virtual reality, augmented reality (AR), and mixed reality. Free viewpoint video (FVV), due to its applications in remote surveillance, remote education, and so on, based on the flexible selection of direction and viewpoint, has been perceived as the development direction of next-generation video technologies and has drawn a wide range of researchers' attention. Since FVV images are synthesized via a depth image-based rendering (DIBR) procedure in the "blind" environment (without reference images), a reliable real-time blind quality evaluation and monitoring system is urgently required. But existing assessment metrics do not render human judgments faithfully mainly because geometric distortions are generated by DIBR. To this end, this paper proposes a novel referenceless quality metric of DIBR-synthesized images using the autoregression (AR)-based local image description. It was found that, after the AR prediction, the reconstructed error between a DIBR-synthesized image and its AR-predicted image can accurately capture the geometry distortion. The visual saliency is then leveraged to modify the proposed blind quality metric to a sizable margin. Experiments validate the superiority of our no-reference quality method as compared with prevailing full-, reduced-, and no-reference models. |
author2 |
School of Computer Science and Engineering |
author_facet |
School of Computer Science and Engineering Gu, Ke Jakhetiya, Vinit Qiao, Jun-Fei Li, Xiaoli Lin, Weisi Thalmann, Daniel |
format |
Article |
author |
Gu, Ke Jakhetiya, Vinit Qiao, Jun-Fei Li, Xiaoli Lin, Weisi Thalmann, Daniel |
author_sort |
Gu, Ke |
title |
Model-based referenceless quality metric of 3D synthesized images using local image description |
title_short |
Model-based referenceless quality metric of 3D synthesized images using local image description |
title_full |
Model-based referenceless quality metric of 3D synthesized images using local image description |
title_fullStr |
Model-based referenceless quality metric of 3D synthesized images using local image description |
title_full_unstemmed |
Model-based referenceless quality metric of 3D synthesized images using local image description |
title_sort |
model-based referenceless quality metric of 3d synthesized images using local image description |
publishDate |
2020 |
url |
https://hdl.handle.net/10356/142320 |
_version_ |
1681056060571385856 |