LOCAL AUGMENTED REALITY PERSISTENT STICKER OBJECTS

• US 20170221272A1
• Filed: 01/29/2016
• Published: 08/03/2017
• Est. Priority Date: 01/29/2016
• Status: Active Grant

• Current Assignee: Snap, Inc.
• Original Assignee: Linjie Luo
• Inventors: Li, Jia, Luo, Linjie, Sheth, Rahul, Xu, Ning, Yang, Jianchao
• Granted Patent: US 10,055,895 B2
• Time in Patent Office: Days
• CPC Class Codes: G06T 19/006 Mixed reality object pose determination, tracking or camera calibration for mixed reality G06T7/00
• CPC Class Codes: G06T 19/20 Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
• CPC Class Codes: G06T 2207/20104 Interactive definition of region of interest [ROI]
• CPC Class Codes: G06T 7/248 involving reference images or patches
• CPC Class Codes: H04L 51/10 Multimedia information
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

Systems and methods for local augmented reality (AR) tracking of an AR object are disclosed. In one example embodiment a device captures a series of video image frames. A user input is received at the device associating a first portion of a first image of the video image frames with an AR sticker object and a target. A first target template is generated to track the target across frames of the video image frames. In some embodiments, global tracking based on a determination that the target is outside a boundary area is used. The global tracking comprises using a global tracking template for tracking movement in the video image frames captured following the determination that the target is outside the boundary area. When the global tracking determines that the target is within the boundary area, local tracking is resumed along with presentation of the AR sticker object on an output display of the device.

Claims

1. A method for local augmented reality (AR) tracking, the method comprising:

capturing, using an image sensor and one or more processors of a device, a first plurality of images of a scene;

displaying the first plurality of images on a display of the device;

receiving, at an input component of the device, a first user selection of an AR sticker object;

receiving, at the input component of the device, a second user selection placing the AR sticker object relative to a first image of the first plurality of images as displayed on the display of the device;

processing, using the one or more processors, one or more images of the first plurality of images to generate a local AR model of the scene; and

adding the AR sticker object to the local AR model of the scene for local tracking of the AR sticker object and presentation of the AR sticker object with AR images on the display of the device.

2. The method of claim 1 further comprising:

capturing, using the image sensor, a second plurality of images of the scene; and

generating, using the one or more processors, a second plurality of AR images using the local AR model of the scene following addition of the AR sticker object to the local AR model of the scene.

3. The method of claim 2 further comprising:

displaying a first AR image of the plurality of AR images, wherein the first AR image comprises the AR sticker object;

displaying a second AR image of the plurality of AR images following display of the first AR image of the plurality of AR images, wherein the second AR image does not comprise the AR sticker object, based on a first movement of the image sensor away from a portion of the scene associated with the AR sticker object; and

displaying a third AR image of the plurality of AR images following display of the second AR image of the plurality of AR images, wherein the third AR image comprises the AR sticker object based on a second movement of the image sensor toward the portion of the scene associated with the AR sticker object.

4. The method of claim 2 further comprising:

displaying a first AR image of the plurality of AR images, wherein the first AR image comprises the AR sticker object;

displaying a second AR image of the plurality of AR images following display of the first AR image of the plurality of AR images, wherein the second AR image does not comprise the AR sticker object, based on a first change in the scene that results in an AR target object moving out of a field of view of the image sensor; and

displaying a third AR image of the plurality of AR images following display of the second AR image of the plurality of AR images, wherein the third AR image comprises the AR sticker object based on a second change in the scene that results in the AR target object moving into the field of view of the image sensor.

5. The method of claim 1 wherein the local AR model is generated in response to the placement of the AR sticker object.

6. The method of claim 1 wherein processing the one or more images of the first plurality of images to generate the local AR model of the scene comprises:

identifying, based on the placement of the AR sticker object, a target within the first image of the first plurality of images; and

tracking movement of the target within subsequent images captured by the image sensor.

7. The method of claim 6 wherein processing the one or more images of the first plurality of images to generate the local AR model of the scene further comprises:

processing a second image of the subsequent images to determine that the target is outside of a boundary area; and

based on the determination that the target is outside of the boundary area, for a third plurality of images captured by the image sensor after the determination that the target is outside the boundary area;

for each image of the third plurality of images, sampling a set of global image points; and

determining an associated movement.

8. The method of claim 7 further comprising:

determining, for a final image of the third plurality of images based on the associated movement for the final image and the sampling of the set of global image points for the final image of the third plurality of images, that the target is within the boundary; and

based on the determination that the target is within the boundary, tracking the target within a fourth plurality of images that follow the third plurality of images.

9. The method of claim 1, wherein adding the AR sticker object to the local AR model of the scene for local tracking of the AR sticker object comprises:

identifying a target template associated with the AR sticker object described by J_i=I₀ (s_i);

wherein J_i is the target template associated with the AR sticker object, the target template comprising a set of color values sampled at a plurality of sample points S_i associated with a target and a target area in an initial AR model image I₀ of the one or more images of the first plurality of images.

10. The method of claim 9 further comprising tracking the AR sticker object by:

processing a second plurality of images I_n received after the initial AR model image such that;

J_i^(l)=I_n(T(s_i))

where T(n) is a transform describing movement of the target between sequential images of the second plurality of images analyzed by the one or more processors; and

calculating a local tracking energy to determine an updated target template for each image of the second plurality of images according to an iterative nonlinear optimization;

11. The method of claim 10, further comprising:

determining a pixel margin for the target;

determining when a center of the target has moved from inside the pixel margin to outside the pixel margin; and

initiating a global movement tracking of the scene for a third plurality of images following the second plurality of images based on the determination that the center of the target has moved outside the pixel margin.

12. The method of claim 11 wherein the center of the target is identified by a center pixel of the target template moving to within a set number of pixels from an edge of a trigger image, wherein the set number of pixels is associated with the pixel margin.

13. The method of claim 12 wherein the global movement tracking comprises:

sampling a fixed set of points g within each image of the third plurality of images following the initiation of the global movement tracking; and

calculating a transform AT to estimate a minimum value for;

14. The method of claim 13 wherein the minimum value is estimated using:

15. A device comprising:

a display;

an input component coupled to the display;

a memory coupled to the display and the input component;

an image sensor; and

one or more processors coupled to the display, the image sensor, the input component, and the memory, the one or more processors configured to process video image frames captured by the image sensor and output local AR images using local AR tracking of an AR sticker object by;

processing a user input associating a first portion of a first image of the video image frames with the AR sticker object and a target;

generating, based on the user input and the first portion of the first image, a first target template associated with the target;

tracking the target across frames of the video image frames following the first image by calculating changes in the first portion of the first image using the first target template;

initiating a global tracking based on a determination that the target is outside a boundary area the global tracking comprising using a global tracking template for tracking movement in the video image frames captured following the determination that the target is outside the boundary area; and

resuming tracking the target when the global tracking determines that the target is within the boundary area, and displaying the AR sticker object on the display based on the tracking of the target.

16. The device of claim 15, wherein tracking the target across the frames of the video image frames following the first image by calculating the changes in the first portion of the first image using the first target template comprises:

identifying the first target template associated with the AR sticker object described by J_i=I₀ (s_i);

wherein J_i is the first target template associated with the AR sticker object, the first target template comprising a set of color values sampled at a plurality of sample points S_i associated with the target and a target area in an initial AR model image I₀ of one or more images of the first plurality of images.

17. The device of claim 16 further comprising tracking the AR sticker object by:

processing a second plurality of images I_n received after the initial AR model image such that;

J_i^(l)=I_n(T(s_i))

where T(n) is a transform describing movement of the target between sequential images of the second plurality of images analyzed by the one or more processors; and

calculating a local tracking energy to determine an updated target template for each image of the second plurality of images according to an iterative nonlinear optimization;

18. A non-transitory computer readable medium comprising instructions that, when performed by one or more processors of a device, cause the device to perform a method comprising:

processing a user input associating a first portion of a first image of video image frames captured by an image sensor with an AR sticker object and a target;

generating, based on the user input and the first portion of the first image, a first target template associated with the target; and

tracking the target across frames of the video image frames following the first image by calculating changes in the first portion of the first image using the first target template.

19. The non-transitory computer readable medium of claim 18 wherein the method performed by the one or more processors of the device further comprises:

initiating a global tracking based on a determination that the target is outside a boundary area the global tracking comprising using a global tracking template for tracking movement in the video image frames captured following the determination that the target is outside the boundary area; and

resuming tracking the target when the global tracking determines that the target is within the boundary area.

20. The non-transitory computer readable medium of claim 19 wherein tracking the target across the frames of the video image frames following the first image by calculating the changes in the first portion of the first image using the first target template comprises:

identifying the first target template associated with the AR sticker object described by J_i=I₀ (s_i);

wherein J_i is the first target template associated with the AR sticker object, the first target template comprising a set of color values sampled at a plurality of sample points S_i associated with the target and a target area in an initial AR model image I₀ of one or more images of the first plurality of images;

tracking the AR sticker object by processing a second plurality of images I_n received after the initial AR model image such that;

J_i^(l)=I_n(T(s_i))

where T(n) is a transform describing movement of the target between sequential images of the second plurality of images analyzed by the one or more processors; and

calculating a local tracking energy to determine an updated target template for each image of the second plurality of images according to an iterative nonlinear optimization;