IP Library Granted Patent US 11,756,173
Granted Patent B2
US 11,756,173 · App. 17/245,824 · Granted Sep 12, 2023

Real-time video enhancement and metadata sharing

Inventor: Reza Rassool (Seattle, WA)
Assignee: REALNETWORKS LLC
G06T5/009G06T5/10G06T2207/10016G06T2207/20208
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 11,756,173
App. No.
17/245,824
Granted
Sep 12, 2023
Kind
B2
Abstract

Embodiments are directed towards video enhancement in accordance with time constraint. An example method includes determining a time constraint for transforming low dynamic range (LDR) video content to high dynamic range (HDR) video content, processing the LDR video content to generate instructions for transforming to HDR video content in accordance with the time constraint, rendering the HDR video content based on executing the generated instructions; and producing metadata including the generated instructions for sharing.

Claims (43)

1. A computing device, comprising:

memory that stores computer instructions; and

at least one processor that executes the computer instructions to perform actions, the actions comprising:

determining a time constraint for transforming, via the computing device, target low dynamic range (LDR) video content to high dynamic range (HDR) video content;

receiving the target LDR video content for presentation to a user as the HDR video content;

processing the target LDR video content to generate instructions for transforming the target LDR video content to the HDR video content in accordance with the time constraint, including:

selecting or configuring a machine learning model, including establishing a reduced number of neural network layers, based, at least in part, on the time constraint; and

using, in real time as the target LDR video content is received, the machine learning model to generate inverse tone map (ITM) parameters and an ITM for the target LDR video content;

applying the generated instructions, including the ITM parameters and the ITM, in real time to the target LDR video content to generate the HDR video content;

rendering the HDR video content to the user in real time as the target LDR video content is received and as the HDR video content is generated; and

producing metadata that includes the generated instructions based on the ITM parameters and the ITM for sharing with one or more other computing devices.

2. The computing device of claim 1 , wherein the HDR video content is superior to the target LDR video content in at least one of resolution, frame rate, or color dynamic range.

3. The computing device of claim 1 , wherein the time constraint corresponds to a threshold of time delay between receiving a stream of the target LDR video content and concurrently rendering a stream of the HDR video content.

4. The computing device of claim 1 , wherein selecting or configuring the machine learning model based, at least in part, on the time constraint comprises establishing a simplified model structure based on the time constraint.

5. The computing device of claim 1 , wherein the actions further comprise determining whether metadata applicable to instruct the computing device to transform the target LDR video content to HDR video content is obtained from an external service.

6. The computing device of claim 5 , wherein the processing of the target LDR video content to generate instructions is performed in response to determining that the metadata is not obtained.

7. The computing device of claim 5 , wherein the actions further comprise providing the produced metadata to the external service.

8. A method, comprising:

determining, by a user device, a time constraint for enhancing target video content;

receiving, by the user device, the target video content for presentation to a user as enhanced video content;

processing, by the user device, the target video content to generate instructions for transforming the target video content into the enhanced video content in accordance with the time constraint, including:

selecting or configuring, by the user device, a machine learning model, including establishing a reduced number of neural network layers, based, at least in part, on the time constraint; and

using, by the user device in real time as the target video content is received, the machine learning model to generate inverse tone map (ITM) parameters and an ITM for the target video content;

applying, by the user device, the generated instructions, including the ITM parameters and the ITM, in real time to the target video content to generate the enhanced video content;

rendering, by the user device in real time as the target video content is received and the enhanced video content is generated, the enhanced video content based, at least in part, on executing the generated instructions; and

producing, by the user device, metadata that includes the generated instructions based on the ITM parameters and the ITM for sharing with one or more other user devices.

9. The method of claim 8 , wherein the time constraint corresponds to a threshold of time delay between receiving a stream of the target video content and concurrently rendering a stream of the enhanced video content.

10. The method of claim 8 , wherein selecting Or configuring the machine learning model based, at least in part, on the time constraint comprises establishing a simplified model structure based on the time constraint.

11. The method of claim 8 , further comprising determining whether metadata applicable to instruct the user device to enhance the target video content is obtained from an external service.

12. The method of claim 11 , wherein the processing of the target video content to generate instructions is performed in response to determining that the metadata is not obtained.

13. The method of claim 11 , further comprising providing the produced metadata to the external service.

14. A non-transitory computer-readable storage medium storing contents that, when executed by one or more processors, cause the one or more processors to:

determine a time constraint for enhancing, via a user device, target video content;

receive the target video content for presentation to a user of the user device as enhanced video content;

process the target video content to generate instructions for enhancing the target video content into the enhanced video content in accordance with the time constraint, including:

select or configure a machine learning model, including establishing a reduced number of neural network layers, based, at least in part, on the time constraint; and

use, in real time as the target video content is received, the machine learning model to generate inverse tone map (ITM) parameters and an ITM for the target video content;

apply the generated instructions, including the ITM parameters and the ITM, in real time to the target video content to generate the enhanced video content;

render, in real time as the target video content is received and the enhanced video content is generated, the enhanced video content based, at least in part, on executing the generated instructions; and

produce metadata that includes the generated instructions based on the ITM parameters and the ITM for sharing with one or more other user devices.

15. The computer-readable storage medium of claim 14 , wherein the time constraint corresponds to a threshold of time delay between receiving a stream of the target video content and concurrently rendering a stream of the enhanced video content.

16. The computer-readable storage medium of claim 14 , wherein the generated instructions include parameters for computing an inverse tone map (ITM).

17. The computer-readable storage medium of claim 14 , wherein the contents further cause the one or more processors to share the produced metadata with the one or more other user devices via an external service.

Assignments (2)
MERGER AND CHANGE OF NAME Recorded Feb 13, 2023
From: REALNETWORKS, INC.; GREATER HEIGHTS ACQUISITION LLC
To: REALNETWORKS LLC
Reel/Frame 062746/0554 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 15, 2021
From: RASSOOL, REZA
To: REALNETWORKS, INC.
Reel/Frame 057806/0707 →
Continuity (1)
Related Publication 20220351342A1 · Nov 3, 2022