Assessing consistency between facial motion and speech signals in video
AssigneeIBM Corporation
InventorsGiridharan Iyengar, Chalapathy V. Neti, Harriet J. Nock
Life of this patent
Estimating the term…
Abstract
The use of multiple complementary classes of measure to assess face and speech consistency in video. In an exemplary embodiment, both synchrony measures and plausibility measures are employed.
Claims23
An apparatus for determining consistency between facial movements and at least one speech signal associated with video, said apparatus comprising:
- an arrangement for extracting a video signal from a video input;
- an arrangement for discerning facial movements in the video signal;
- an arrangement for obtaining at least one speech signal associated with the video input;
- an arrangement for producing a synchrony measure which measures consistency between the facial movements discerned and the at least one speech signal obtained;
- and
- an arrangement for producing a plausibility measure which measures consistency between the facial movements discerned at the at least one speech signal obtained.
A method of determining consistency between facial movements and at least one speech signal associated with video, said method comprising the steps of:
- extracting a video signal from a video input;
- discerning facial movements in the video signal;
- obtaining at least one speech signal associated with the video input;
- producing a synchrony measure which measures consistency between the facial movements discerned and the at least one speech signal obtained;
- and
- producing a plausibility measure which measures consistency between the facial movements discerned at the at least one speech signal obtained.
A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps determining consistency between facial movements and at least one speech signal associated with video, said method comprising the steps of:
- extracting a video signal from a video input;
- discerning facial movements in the video signal;
- obtaining at least one speech signal associated with the video input;
- producing a synchrony measure which measures consistency between the facial movements discerned and the at least one speech signal obtained;
- and
- producing a plausibility measure which measures consistency between the facial movements discerned at the at least one speech signal obtained.
Dependencies are inferred from the claim text.