Image Processor with Static Hand Pose Recognition Utilizing Contour Triangulation and Flattening

• US 20150278589A1
• Filed: 03/25/2015
• Published: 10/01/2015
• Est. Priority Date: 03/27/2014
• Status: Abandoned Application

• Current Assignee: Avago Technologies General IP PTE Limited (Broadcom, Inc.)
• Original Assignee: Avago Technologies General IP PTE Limited (Broadcom, Inc.)
• Inventors: Mazurenko, Ivan Leonidovich, Kholodenko, Alexander Borisovich, Babin, Dmitry Nicolaevich, Letunovskiy, Aleksey Alexandrovich, Petyushko, Alexander Alexandrovich
• Application Number:
  US14/668,625
• Publication Number:
  US 20150278589A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G01C 11/30 by triangulation
• CPC Class Codes: G06V 20/64 Three-dimensional objects
• CPC Class Codes: G06V 40/107 Static hand or arm

Abstract

An image processing system comprises an image processor having image processing circuitry and an associated memory. The image processor is configured to implement a gesture recognition system utilizing the image processing circuitry and the memory. The gesture recognition system implemented by the image processor comprises a static pose recognition module. The static pose recognition module is configured to identify a hand region of interest in at least one image, to determine a contour of the hand region of interest, to triangulate the determined contour, to flatten the triangulated contour, to compute one or more features of the flattened contour, and to recognize a static pose of the hand region of interest based at least in part on the one or more computed features.

Claims

1. A method comprising steps of:

identifying a hand region of interest in at least one image;

determining a contour of the hand region of interest;

triangulating the determined contour;

flattening the triangulated contour;

computing one or more features of the flattened contour; and

recognizing a static pose of the hand region of interest based at least in part on the one or more computed features;

wherein the steps are implemented in an image processor comprising a processor coupled to a memory.

2. The method of claim 1 wherein the steps are implemented in a static pose recognition module of a gesture recognition system of the image processor.

3. The method of claim 1 wherein identifying a hand region of interest comprises generating a hand image comprising a binary region of interest mask in which pixels within the hand region of interest all have a first binary value and pixels outside the hand region of interest all have a second binary value complementary to the first binary value.

4. The method of claim 1 wherein the determined contour comprises an ordered list of points.

5. The method of claim 1 wherein triangulating the determined contour comprises covering substantially all of an area bounded by the determined contour using triangles with vertices that correspond to respective contour points.

6. The method of claim 5 wherein all of the vertices of the triangles used in the triangulating step correspond to contour points such that none of the triangles includes a vertex inside the bounded area.

7. The method of claim 1 wherein flattening the triangulated contour comprises altering one or more angles between respective pairs of triangles in the triangulated contour.

8. The method of claim 7 wherein altering one or more angles between respective pairs of triangles comprises replacing each of a plurality of such angles with an angle of approximately 180 degrees.

9. The method of claim 1 wherein determining a contour of the hand region of interest comprises:

determining a two-dimensional contour of the hand region of interest; and

converting the two-dimensional contour to a three-dimensional contour;

wherein the triangulating step is applied to the three-dimensional contour.

10. The method of claim 9 further comprising applying at least one of a simplification operation and a smoothing operation to the two-dimensional contour prior to converting the two-dimensional contour to a three-dimensional contour.

11. The method of claim 9 further comprising:

identifying a palm boundary of the three-dimensional contour.

12. The method of claim 11 further comprising:

modifying the three-dimensional contour to exclude contour points outside the identified palm boundary.

13. The method of claim 9 further comprising regularizing the three-dimensional contour by altering at least one of number and distribution of its contour points prior to applying the triangulating step to the three-dimensional contour.

14. The method of claim 9 wherein converting the two-dimensional contour to a three-dimensional contour utilizes a refined depth map comprising one or more reconstructed depth values generated for respective pixels.

15. The method of claim 14 wherein a given one of the reconstructed depth values is generated at least in part based on one or more pixels of an input depth map that are not part of the hand region of interest but are instead within a specified neighborhood of at least one pixel that is part of the hand region of interest.

16. An apparatus comprising:

an image processor comprising image processing circuitry and an associated memory;

wherein the image processor is configured to implement a gesture recognition system utilizing the image processing circuitry and the memory, the gesture recognition system comprising a static pose recognition module; and

wherein the static pose recognition module is configured to identify a hand region of interest in at least one image, to determine a contour of the hand region of interest, to triangulate the determined contour, to flatten the triangulated contour, to compute one or more features of the flattened contour, and to recognize a static pose of the hand region of interest based at least in part on the one or more computed features.

17. The apparatus of claim 16 wherein the determined contour comprises an ordered list of points.

18. The apparatus of claim 16 wherein the triangulated contour comprises covering substantially all of an area bounded by the determined contour using triangles with vertices that correspond to respective contour points.

19. The apparatus of claim 18, wherein all of the vertices of the triangles used in the triangulated contour correspond to contour points such that none of the triangles includes a vertex inside the bounded area.

20. The apparatus of claim 16 wherein the static pose recognition module determines the contour of the hand region of interest by determining a two-dimensional contour of the hand region of interest and converting the two-dimensional contour to a three-dimensional contour, and wherein the three-dimensional contour is triangulated to provide the triangulated contour.