Three dimensional close interactions

• US 9,189,855 B2
• Filed: 06/06/2013
• Issued: 11/17/2015
• Est. Priority Date: 06/25/2012
• Status: Active Grant

• Current Assignee: SoftKinetic Software
• Original Assignee: SoftKinetic Software
• Inventors: Guigues, Laurent, Simons, Kevin, Hou, Yunshu, Feys, Olivier, Kamovich, Aliaksandr
• Primary Examiner(s): STREGE, JOHN B
• Application Number:
  US14/366,640
• Publication Number:
  US 20150117708A1
• Time in Patent Office: 894 Days
• Field of Search: 382/103, 382/173, 382/190, 382/195
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/10028 Range image; Depth image; 3D point clouds
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments
• CPC Class Codes: G06T 7/73 using feature-based methods

Abstract

Described herein is a method for detecting, identifying and tracking hand, hand parts, and fingers on the hand (500) of a user within depth images of a three-dimensional scene. Arms of a user are detected, identified, segmented from the background of the depth images, and tracked with respect to time. Hands of the user are identified and tracked and the location and orientation of its parts, including the palm and the fingers (510, 520, 530, 540, 550) are determined and tracked in order to produce output information enabling gesture interactions.

Claims

1. A method of detecting and tracking at least one feature associated with an extremity of a user in a three-dimensional scene using depth images of the scene, the method comprising the steps of:

a) segmenting at least one region of interest within each depth image;

b) determining at least one object of interest that corresponds to a limb of the user within the segmented depth image;

c) tracking each limb with respect to time;

d) identifying an extremity associated with each limb;

e) tracking each identified extremity with respect to time;

isolating each identified extremity from its associated limb;

g) determining parameters relating to each isolated extremity;

h) identifying at least one feature in each isolated extremity; and

i) tracking each identified feature with respect to time.

2. A method according to claim 1, wherein step a) comprises utilising at least one structuring element adapted to be a function of the depth of each pixel in the depth image.

3. A method according to claim 1, wherein step c) comprises determining temporal associations for each limb.

4. A method according to claim 3, wherein step e) comprises tracking each extremity based on temporal associations assessed for each limb.

5. A method according to claim 1, wherein step d) further comprises using a weighted average of points of each limb in the depth image to identify each extremity.

6. A method according to claim 1, wherein step f) comprises the steps of:

f1) determining a centre of each extremity associated with each limb of the user; and

f2) determining a distance from the centre of the extremity to its boundary.

7. A method according to claim 6, wherein step f) further comprises using a cutting process to isolate the extremity from the limb, the cutting process being centred on the determined centre of the extremity.

8. A method according to claim 6, wherein step f) further comprises using a growing process to isolate the extremity from the limb.

9. A method according to claim 6, wherein step g) comprises the step of determining at least one parameter associated with the extremity.

10. A method according to claim 9, wherein step g) further comprises the step of using each determined parameter associated with the extremity to identify at least one feature associated with the extremity.

11. A method according to claim 9, wherein the at least one parameter associated with the extremity comprises the orientation of the extremity within the depth image.

12. A method according to claim 9, wherein step g) further comprises using each determined parameter to determine a tip position of the extremity.

13. A method according to claim 9, wherein the limb comprises an arm and the extremity comprises a hand, and wherein each feature associated with the extremity comprises at least one finger of the hand.

14. A method according to claim 13, wherein step h) comprises identifying silhouette fingers and pointing fingers within a hand contour.

15. A method according to claim 14, wherein step h) comprises merging the identified silhouette fingers and pointing fingers to form a unique set of fingers.

16. A gesture recognition method for recognising gestures made by at least one hand, comprising the method of detecting and tracking at least one feature associated with an extremity of a user in a three-dimensional scene using depth images of the scene according to claim 13.

17. A non-transitory computer readable medium storing a program causing a computer to execute a method for performing the steps of the method of detecting and tracking at least one feature associated with an extremity of a user in a three-dimensional scene using depth images of the scene, the method comprising the steps of:

a) segmenting at least one region of interest within each depth image;

b) determining at least one object of interest that corresponds to a limb of the user within the segmented depth image;

c) tracking each limb with respect to time;

d) identifying an extremity associated with each limb;

e) tracking each identified extremity with respect to time;

f) isolating each identified extremity from its associated limb;

g) determining parameters relating to each isolated extremity;

h) identifying at least one feature in each isolated extremity; and

i) tracking each identified feature with respect to time.

18. A gesture recognition system for detecting and tracking at least one feature associated with an extremity of a user in a three-dimensional scene using depth images of the scene, the gesture recognition system including a computerised system capable of performing the following steps:

a) segmenting at least one region of interest within each depth image;

b) determining at least one object of interest that corresponds to a limb of the user within the segmented depth image;

c) tracking each limb with respect to time;

d) identifying an extremity associated with each limb;

e) tracking each identified extremity with respect to time;

f) isolating each identified extremity from its associated limb;

g) determining parameters relating to each isolated extremity;

h) identifying at least one feature in each isolated extremity; and

i) tracking each identified feature with respect to time.