초록 ▼

Visual images of the face of a speaker are processed to extract during a learning sequence a still frame of the image and a set of typical mouth shapes. Encoding of a sequence to be transmitted, recorded etc. is then achieved by matching the changing mouth shapes to those of the set and generating c

Visual images of the face of a speaker are processed to extract during a learning sequence a still frame of the image and a set of typical mouth shapes. Encoding of a sequence to be transmitted, recorded etc. is then achieved by matching the changing mouth shapes to those of the set and generating codewords identifying them. Alternatively, the codewords may be generated to accompany real or synthetic speech using a look- up table relating speech parameters to codewords. In a receiver, the still frames and set of mouth shapes are stored and received codewords used to select successive mouth shapes to be incorporated in the still frame.

대표청구항 ▼

An apparatus for encoding a moving image including a human face comprising: means for receiving video input data; means for output of data representing one frame of the image; identification means arranged in operation for each frame of the image to identify that part of the input data corresponding

An apparatus for encoding a moving image including a human face comprising: means for receiving video input data; means for output of data representing one frame of the image; identification means arranged in operation for each frame of the image to identify that part of the input data corresponding to the mouth of the face represented and (a) in a first phase of operation to compare the mouth data parts of each frame with those of other frames to select a representative set of mouth data parts, to store the selected parts and to output the selected parts; (b) in a second phase to compare the mouth data part of each frame with those of the stored set and to generate a codeword indicating which member of the set the mouth data part of that frame most closely resembles.