摘要:
A (50) method and apparatus (110) for determining a scanning format of a frame of a video signal determines (52) if any interfield motion exists within the frame; and indicates (53) that the frame is interlaced scanned if interfield motion is detected and indicates (54) that the frame is progressive scanned if interfield motion is not detected. The determination is based on spatial correlation within and around one or more edges of one or more objects in the frame. This can be accomplished by identifying zig-zag or zipper effects near one or more edges of one or more objects in the frame, identifying motion between fields or identifying motion near one or more edges of one or more objects. A method (100) for determining a field order of two interlaced frames of a video sequence of frames assigns (101) one of the two interlaced frames to be a first frame and the other of the two interlaced frames to be a second frame and assembles (102) two new frames from the two interlaced frames by combining a bottom field of the first frame with the top field of the second frame and by combining a top field of the first frame with a bottom field of the second frame. The number of zipper points in each of the two new frames is then determined and compared. If the number of zipper points in the first frame is smaller than the number of zipper points in the second frame, that the field order of the two frames is top field first, otherwise the field order is bottom field first.
摘要:
Field order in a sequence of video frames is determined based on magnitude or relative motion of a motion vector between a bottom field of a current frame and a top field of a next frame, and magnitude or relative motion of a motion vector between a top field of the current frame and a bottom field of the next frame. Field order of two interlaced frames of a video sequence is determined by assembling two new frames from the two interlaced frames by combining a bottom field of the first frame with a top field of the second frame and by combining a top field of the first frame with a bottom field of the second frame. The field order is top field first if the number of zipper points in the first new frame is smaller than the number of zipper points in the second new frame.
摘要:
A method and apparatus is provided for annotating video content with metadata generated using speech recognition technology. The method begins by rendering video content on a display device. A segment of speech is received from a user such that the speech segment annotates a portion of the video content currently being rendered. The speech segment is converted to a text-segment and the text-segment is associated with the rendered portion of the video content. The text segment is stored in a selectively retrievable manner so that it is associated with the rendered portion of the video content.
摘要:
A method and apparatus is provided for annotating video content with metadata generated using speech recognition technology. The method begins by rendering video content on a display device. A segment of speech is received from a user such that the speech segment annotates a portion of the video content currently being rendered. The speech segment is converted to a text-segment and the text-segment is associated with the rendered portion of the video content. The text segment is stored in a selectively retrievable manner so that it is associated with the rendered portion of the video content.