Detecting anchor shots accurately is very improtant for automatically parsing news video and extracting meaningful news items. The paper proposes a fast anchor shot detection algorithm,which is based on backgroud chor...Detecting anchor shots accurately is very improtant for automatically parsing news video and extracting meaningful news items. The paper proposes a fast anchor shot detection algorithm,which is based on backgroud choronminance and skin tone models. The attractive characteristics of the algorithminclude only simple computation involved. At the same time,it operates in MPEG compression domain,which makes the detection speed very fast. The algorithm was evaluated on a big test set containing more than 480000 frames and news video from two different TV stations. More than 98% accuracy and 100% recall have been gained. The experiment results also show the system has an average detection speed of 77.55 f/s. The experiments demonstrate the algorithm is a fast and effective one.展开更多
文摘Detecting anchor shots accurately is very improtant for automatically parsing news video and extracting meaningful news items. The paper proposes a fast anchor shot detection algorithm,which is based on backgroud choronminance and skin tone models. The attractive characteristics of the algorithminclude only simple computation involved. At the same time,it operates in MPEG compression domain,which makes the detection speed very fast. The algorithm was evaluated on a big test set containing more than 480000 frames and news video from two different TV stations. More than 98% accuracy and 100% recall have been gained. The experiment results also show the system has an average detection speed of 77.55 f/s. The experiments demonstrate the algorithm is a fast and effective one.
