Viewing Experience Model of First-Person Videos
AbstractFirst-Person Videos (FPVs) are recorded using wearable cameras to share the recorder’s First-Person Experience (FPE). Ideally, the FPE is conveyed by the viewing experience of the FPV. However, raw FPVs are usually too shaky to watch, which ruins the viewing experience. To solve this problem, we improve the viewing experience of FPVs by modeling it as two parts: video stability and First-Person Motion Information (FPMI). Existing video stabilization techniques can improve the video stability but damage the FPMI. We propose a Viewing Experience (VE) score, which measures both the stability and the FPMI of a FPV by exploring the mechanism of human perception. This enables us to further develop a system that can stabilize FPVs while preserving their FPMI so that the viewing experience of FPVs is improved. Objective tests show that our measurement is robust under different kinds of noise, and our system has competitive performance relative to current stabilization techniques. Subjective tests show that (1) both our stability and FPMI measurements can correctly compare the corresponding attributes of an FPV across different versions of the same content, and (2) our video processing system can effectively improve the viewing experience of FPVs. View Full-Text
Share & Cite This Article
Ma, B.; Reibman, A.R. Viewing Experience Model of First-Person Videos. J. Imaging 2018, 4, 106.
Ma B, Reibman AR. Viewing Experience Model of First-Person Videos. Journal of Imaging. 2018; 4(9):106.Chicago/Turabian Style
Ma, Biao; Reibman, Amy R. 2018. "Viewing Experience Model of First-Person Videos." J. Imaging 4, no. 9: 106.
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.