An Error-Activation-Guided Blind Metric for Stitched Panoramic Image Quality Assessment

被引:5
作者
Yang, Luyu [1 ]
Liu, Jiang [2 ]
Gao, Chenqiang [3 ]
机构
[1] Kandao Technol, Shenzhen, Peoples R China
[2] Carnegie Mellon Univ, Pittsburgh, PA 15213 USA
[3] Chongqing Univ Posts & Telecommun, Chongqing Key Lab Signal & Informat Proc, Chongqing, Peoples R China
来源
COMPUTER VISION, PT II | 2017年 / 772卷
基金
中国国家自然科学基金;
关键词
Image quality assessment; Multi-view synthesis; Virtual reality;
D O I
10.1007/978-981-10-7302-1_22
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Image stitching is one key enabling component for recent immersive VR technology. The quality of the stitched images greatly affects VR experiences. Evaluation of stitched panoramic images using existing assessment tools is insufficient for two reasons. First, conventional image quality assessment (IQA) metrics are mostly full-referenced, while panorama reference is hard to obtain. Second, existing IQA metrics are not designed to detect and evaluate errors typical in stitched images. In this paper, we design an IQA metric for stitched images, where ghosting and shape inconsistency are the most common visual distortions. Specifically, we first locate the error with a fine-tuned convolutional neural network (CNN), and later refine the locations using an error-activation mapping generated from the network. Each located error is defined by both its size and distortion level. Extensive experiments and comparisons confirm the effectiveness of our metric, and indicate the network's remarkable ability to detect error patterns.
引用
收藏
页码:256 / 268
页数:13
相关论文
共 26 条
  • [1] Achanta R, 2009, PROC CVPR IEEE, P1597, DOI 10.1109/CVPRW.2009.5206596
  • [2] [Anonymous], ARXIV14093964
  • [3] [Anonymous], 2005, IEEE 7 WORKSH MULT S
  • [4] [Anonymous], 2015, IEEE T IMAGE PROCESS, DOI DOI 10.1109/TIP.2015.2487833
  • [5] [Anonymous], PROC CVPR IEEE
  • [6] Bellver M., 2016, Hierarchical object detection with deep reinforcement learning, V31
  • [7] Automatic panoramic image stitching using invariant features
    Brown, Matthew
    Lowe, David G.
    [J]. INTERNATIONAL JOURNAL OF COMPUTER VISION, 2007, 74 (01) : 59 - 73
  • [8] Full-reference quality assessment of stereopairs accounting for rivalry
    Chen, Ming-Jun
    Su, Che-Chun
    Kwon, Do-Kyoung
    Cormack, Lawrence K.
    Bovik, Alan C.
    [J]. SIGNAL PROCESSING-IMAGE COMMUNICATION, 2013, 28 (09) : 1143 - 1155
  • [9] Dessein A, 2014, IEEE IMAGE PROC, P2031, DOI 10.1109/ICIP.2014.7025407
  • [10] Krizhevsky A., 2017, COMMUN ACM, V60, P84, DOI DOI 10.1145/3065386