NEURAL-NETWORK-BASED DETECTION METHODS FOR COLOR, SHARPNESS, AND GEOMETRY ARTIFACTS IN STEREOSCOPIC AND VR180 VIDEOS

被引：0

作者：

Lavrushkin, Sergey ^{[1
]}

Kozhemyakov, Konstantin ^{[1
]}

Vatolin, Dmitriy ^{[1
]}

机构：

[1] Lomonosov Moscow State Univ, Moscow, Russia

来源：

2020 INTERNATIONAL CONFERENCE ON 3D IMMERSION (IC3D) | 2020年

基金：

俄罗斯基础研究基金会;

关键词：

objective quality assessment; color mismatch; sharpness mismatch; geometric distortions; stereoscopic video; vr180; deep learning;

D O I：

10.1109/IC3D51119.2020.9376385

中图分类号：

TB8 [摄影技术];

学科分类号：

0804 ;

摘要：

Shooting video in 3D format can introduce stereoscopic artifacts, potentially causing viewers visual discomfort. In this work, we consider three common stereoscopic artifacts: color mismatch, sharpness mismatch, and geometric distortion. This paper introduces two neural-network-based methods for simultaneous color- and sharpness-mismatch estimation, as well as for estimating geometric distortions. To train these networks we prepared large datasets based on frames from full-length stereoscopic movies and compared the results with methods that previously served in analyses of full-length stereoscopic movies. We used our proposed methods to analyze 100 videos in VR180 format-a new format for stereoscopic videos in virtual reality (VR). This work presents overall results for these videos along with several examples of detected problems.

引用

页数：8