H. Pan, B. Li, and M. I. Sezan, "Automatic detection of replay segments in broadcast sports programs by detection of logos in scene transitions," Proc. IEEE Int' l . Conf. on Acoustics, Speech, and Signal Processing (ICASSP), 2002.  Home/Search   Document Not in Database   Summary   Related Articles

....cinematographic features in Eqs. 18 20. D. Slow Motion Replay Detection Replays in sports broadcasts are excellent locators of semantically important segments for high level video processing. Several slow motion replay detectors for compressed and spatial domains exist in the literature [25] [27]. Since we only need to determine if a given shot consists of a slow motion segment, the zero crossing measure proposed in [26] has proved to be su#cient for our application. Zero crossing measure evaluates the amplitude of the fluctuations in the frame di#erences (D(t) values) within a ....

....is compensated by #, which defines the threshold for the number of fluctuations in the processed window. Finally, the number of zero crossings, p zc (t) is defined through Eqs. 21 24 as the largest index value, k, for which Z c (t, # k ) #. D(t) I t (i, j) I t 1 (i, j) 21) [27] improves [26] by detecting logo transitions. Since the use of logo transitions before and after replays is broadcaster dependent, 26] is a more generic algorithm. In [26] several other features are also proposed to both locate the slow motions and to find other fields, such as still frames ....

H. Pan, B. Li, and M. I. Sezan, "Automatic detection of replay segments in broadcast sports programs by detection of logos in scene transitions," Proc. IEEE Int' l . Conf. on Acoustics, Speech, and Signal Processing (ICASSP), 2002.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC