Figure - PMC

Skip to main content

View full-text article in PMC

. Author manuscript; available in PMC: 2014 Sep 25.

Published in final edited form as: Proc IEEE Int Conf Comput Vis. 2013;2013:2400–2407. doi: 10.1109/ICCV.2013.298

Example from the RU-FACS dataset [2]: (a) A video of subject 77, (b) Frame detection result in thin orange line and ground truth (GT) in thick gray line. (c) A segment score matrix for frame+segment (FS) detection. The higher the score is, the more likely that there is an AU in this segment. (d) Event score matrix for FS+transition (FST) detection. Using the transition score in (e) as a refinement, FST detector () fires at (12,31), which better approaches the GT. (f) Detected segments.

Inline graphic — Example from the RU-FACS dataset [2]: (a) A video of subject 77, (b) Frame detection result in thin orange line and ground truth (GT) in thick gray line. (c) A segment score matrix for frame+segment (FS) detection. The higher the score is, the more likely that there is an AU in this segment. (d) Event score matrix for FS+transition (FST) detection. Using the transition score in (e) as a refinement, FST detector () fires at (12,31), which better approaches the GT. (f) Detected segments.