Method and system for detecting conscious hand movement patterns and computer-generated visual feedback for facilitating human-computer interaction
The present invention is a system and method for detecting and analyzing motion patterns of individuals present at a multimedia computer terminal from a stream of video frames generated by a video camera and the method of providing visual feedback of the extracted information to aid the interaction process between a user and the system. The method allows multiple people to be present in front of the computer terminal and yet allow one active user to make selections on the computer display. Thus the invention can be used as method for contact-free human-computer interaction in a public place, where the computer terminal can be positioned in a variety of configurations including behind a transparent glass window or at a height or location where the user cannot touch the terminal physically.
1. An apparatus for providing a large number of graphical elements to an individual who is interacting with a multimedia terminal, comprising:
(a) means for capturing a plurality of images from said individual,
(b) means for calculating a color segmentation image of said individual in said plurality of images,
(c) means for calculating a motion energy image of said individual in said plurality of images,
(d) means for calculating a body part location confidence image of said individual by combining said color segmentation image and said motion energy image,
wherein combining said color segmentation image and said motion energy image is performed by adding and multiplying the images together, and
(e) means for displaying said large number of graphical elements on said multimedia terminal that correspond to the body part location estimates of said individual in the body part location confidence image,
whereby said large number of graphical elements provides said individual with information about what the apparatus senses and the certainty with which the apparatus has sensed said body part location estimates,
whereby said large number of graphical elements gives said individual an opportunity to have a better understanding about the capabilities of said multimedia terminal to sense said body movements, and
whereby said understanding helps said individual to adapt to the capabilities of said multimedia terminal, leading to an improved interaction experience.
2. The apparatus according to claim 1 , wherein said means for displaying said large number of graphical elements on said multimedia terminal further comprises means for varying the density of the graphical elements that are displayed in accordance to the values in said body part location confidence image.
3. The apparatus according to claim 1 , wherein said means for calculating said color segmentation image further comprises means for using a statistical color model of said individual's skin color in order to perform the calculation of said color segmentation image.
4. The apparatus according to claim 1 , wherein said means for calculating said motion energy image further comprises means for calculating differences between consecutive images in order to perform the calculation of said motion energy image.
5. The apparatus according to claim 1 , wherein the apparatus further comprises means for allowing said individual to use his or her body movements to interact with said multimedia terminal without physically coming in contact with said terminal, wherein said large number of graphical elements shows said individual a high density at the location to which said individual is pointing.
6. A method for providing a large number of graphical elements to an individual who is interacting with a multimedia terminal, comprising the following steps of:
(a) capturing a plurality of images from said individual,
(b) calculating a color segmentation image of said individual in said plurality of images,
(c) calculating a motion energy image of said individual in said plurality of images,
(d) calculating a body part location confidence image of said individual by combining said color segmentation image and said motion energy image,
wherein combining said color segmentation image and said motion energy image is performed by adding and multiplying the images together, and
(e) displaying said large number of graphical elements on said multimedia terminal that correspond to the body part location estimates of said individual in the body part location confidence image,
whereby said large number of graphical elements provides said individual with information about the certainty of sensed said body part location estimates,
whereby displaying said large number of graphical elements gives said individual an opportunity to have a better understanding about the capabilities of said multimedia terminal, and
whereby said understanding helps said individual to adapt to the capabilities of said multimedia terminal, leading to an improved interaction experience.
7. The method according to claim 6 , wherein the method further comprises a step of varying the density of the graphical elements that are displayed in accordance to the values in said body part location confidence image.
8. The method according to claim 6 , wherein the step of calculating said color segmentation image further comprises a step of using a statistical color model of said individual's skin color in order to perform the calculation of said color segmentation image.
9. The method according to claim 6 , wherein the step of calculating said motion energy image further comprises a step of calculating differences between consecutive images in order to perform the calculation of said motion energy image.
10. The method according to claim 6 , wherein the method further comprises a step of allowing said individual to use his or her body movements to interact with said multimedia terminal without physically coming in contact with said terminal,
wherein said large number of graphical elements shows to said individual a high density at the location to which said individual is pointing.