Title :
Packet Loss Visibility of View+Depth Compressed Stereo 3D Video
Author :
Qing Song ; Cosman, Pamela
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of California, San Diego, La Jolla, CA, USA
Abstract :
We conduct a human observer experiment on the visibility of fixed-sized packet losses in compressed 3D videos. The stereo videos are encoded using view+depth based on H.264/AVC. A prediction model of the visibility score is built with features extracted from the video. Results show good accuracy of the model.
Keywords :
data compression; feature extraction; stereo image processing; video codecs; video coding; H.264-AVC; feature extraction; fixed-sized packet losses; human observer experiment; prediction model; view+depth compressed stereo 3D video; visibility score; Color; Feature extraction; Image color analysis; Packet loss; Three-dimensional displays; Vectors;
Conference_Titel :
Packet Video Workshop (PV), 2013 20th International
Conference_Location :
San Jose, CA
DOI :
10.1109/PV.2013.6691453
