IP Library › Granted Patent US 10,728,568
Granted Patent B1
US 10,728,568 · App. 15/933,028 · Granted Jul 28, 2020

Visual element encoding parameter tuning

Inventors: Charles Benjamin Franklin Waggoner (Portland, OR); Yongjun Wu (Bellevue, WA); Hai Wei (Seattle, WA)
Assignee: Amazon Technologies, Inc.
H04N19/46H04N19/176H04N19/52
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,728,568
App. No.
15/933,028
Granted
Jul 28, 2020
Kind
B1
Abstract

Techniques are described for adaptive encoding of different visual elements in a video frame. Characteristics of visual elements can be determined and used to set encoding parameters for the visual elements. The visual elements can be encoded such that one visual element is encoded differently than another visual element if they have different characteristics.

Claims (38)

1. A computer-implemented method, comprising:

generating a first video frame by integrating a synthetic visual element with a natural image, the first video frame including a plurality of pixels, a first subset of pixels corresponding to fewer than all of the pixels of the first video frame, the first subset of pixels corresponding to the synthetic visual element integrated with the natural image of the first video frame;

generating first metadata based on one or more characteristics of the synthetic visual element, the first metadata representing the one or more characteristics of the synthetic visual element corresponding to the first subset of pixels;

configuring a video encoding parameter for the first subset of pixels based on the first metadata; and

encoding the first video frame using the video encoding parameter.

2. The method of claim 1 , wherein the first metadata represent one or more of a color, position, motion, a shape, a size, scaling, repetition, a transition, a quality, a type, or a level of interest for the synthetic visual element represented by the first subset of pixels.

3. The method of claim 1 , wherein the video encoding parameter corresponds to one or more of motion prediction, block size, predicted block, skip block, sub-pel precision, asymmetric motion partition, transform skip, lossless coding, color transform, adaptive motion vector resolution, intra block copying, color based prediction, quantization parameter, or frame type parameter.

4. The method of claim 1 , wherein the first video frame corresponds to a live broadcast of a live event, and wherein the encoding of the first video frame occurs in near real time relative to the live event.

5. The method of claim 1 , wherein the first metadata for the first subset of pixels also represents a second subset of pixels of a second video frame, the method further comprising encoding the second subset of pixels with reference to the encoding of the first subset of pixels in the first video frame.

6. The method of claim 1 , wherein the first metadata are included in extensible markup language (XML) data, the XML data including timing information, the method further comprising associating the first metadata with the first video frame based on the timing information.

7. The method of claim 1 , further comprising:

analyzing the first video frame to identify the synthetic visual element; and

generating the first metadata based on the analysis of the first video frame.

8. The method of claim 1 , wherein encoding the first video frame includes encoding a first instance and a second instance of the synthetic visual element without performing a motion search, the first instance and the second instance being in different positions in the first video frame.

9. A computer-implemented method, comprising:

generating a first video frame by integrating a visual element with natural imagery, the first video frame including a plurality of pixels, a first subset of pixels corresponding to fewer than all of the pixels of the first video frame, the first subset of pixels corresponding to the visual element in the first video frame;

generating first metadata based on one or more characteristics of the visual element, the first metadata representing the one or more characteristics of the visual element corresponding to the first subset of pixels;

configuring a video encoding parameter for the first subset of pixels based on the first metadata; and

encoding the first video frame using the video encoding parameter.

10. The method of claim 9 , wherein the first metadata represent one or more of a color, position, motion, a shape, a size, scaling, repetition, a transition, a quality, a type, or a level of interest for the visual element represented by the first subset of pixels.

11. The method of claim 9 , wherein the video encoding parameter corresponds to one or more of motion prediction, block size, predicted block, skip block, sub-pel precision, asymmetric motion partition, transform skip, lossless coding, quantization parameter, color transform, adaptive motion vector resolution, intra block copying, color based prediction, or frame type parameter.

12. The method of claim 9 , wherein the first video frame corresponds to a live broadcast of a live event, and wherein the encoding of the first video frame occurs in near real time relative to the live event.

13. The method of claim 9 , wherein the first metadata for the first subset of pixels also represents a second subset of pixels of a second video frame, the method further comprising encoding the second subset of pixels with reference to the encoding of the first subset of pixels in the first video frame.

14. The method of claim 9 , wherein the first metadata are included in extensible markup language (XML) data, the XML data including timing information, the method further comprising associating the first metadata with the first video frame based on the timing information.

15. The method of claim 9 , further comprising:

analyzing the first video frame to identify the visual element; and

generating the first metadata based on the analysis of the first video frame.

16. The method of claim 9 , wherein encoding the first video frame includes encoding a first instance and a second instance of the visual element without performing a motion search, the first instance and the second instance being in different positions in the first video frame.

17. A system, comprising one or more processors and memory configured to:

generate a first video frame by integrating a visual element with natural imagery, the first video frame including a plurality of pixels, a first subset of pixels corresponding to fewer than all of the pixels of the first video frame, the first subset of pixels corresponding to the visual element in the first video frame;

generate first metadata based on one or more characteristics of the visual element, the first metadata representing the one or more characteristics of the visual element corresponding to the first subset of pixels;

configure a video encoding parameter for the first subset of pixels based on the first metadata; and

encode the first video frame using the video encoding parameter.

18. The system of claim 17 , wherein the first metadata represent one or more of a color, position, motion, a shape, a size, scaling, repetition, a transition, a quality, a type, or a level of interest for the visual element represented by the first subset of pixels.

19. The system of claim 17 , wherein the video encoding parameter corresponds to one or more of motion prediction, block size, predicted block, skip block, sub-pel precision, asymmetric motion partition, transform skip, lossless coding, quantization parameter, color transform, adaptive motion vector resolution, intra block copying, color based prediction, or frame type parameter.

20. The system of claim 17 , the one or more processors and memory further configured to:

analyze the first video frame to identify the visual element; and

generate the first metadata based on the analysis of the first video frame.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 22, 2018
From: WAGGONER, CHARLES BENJAMIN FRANKLIN; WU, YONGJUN; WEI, HAI
To: AMAZON TECHNOLOGIES, INC.
Reel/Frame 045320/0671 →
Cited By (1)
US 12,452,436