Associating video frames with text
Pınar Duygulu, Howard D. Wactlar · 2003
to solve the correspondence problem between video frames and associated text in order to annotate video frames with more reliable labels and descriptions. Visual features extracted from video frames are linked to text that is obtained from the audio transcripts using joint statistics. The results show that using this approach it is possible to have better annotations for video frames that can be later used to improve the performance of text based queries. The proposed approach will be integrated into the Informedia Digital Video Library Project at Carnegie Mellon University. 1.