IP Library Granted Patent US 11,930,063
Granted Patent B2
US 11,930,063 · App. 17/543,377 · Granted Mar 12, 2024

Content completion detection for media content

Inventors: Jonathan Bennett-James (Wales, GB); Bineet Kumar Singh (Karnataka, IN); Nishant Kumar (Karnataka, IN)
Assignee: NAGRAVISION S.A.
H04L65/613
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 11,930,063
App. No.
17/543,377
Granted
Mar 12, 2024
Kind
B2
Abstract

Systems and techniques are described herein for processing media content. For example, a process can include obtaining a first media frame and a second media frame. The process can include generating, using a first change detector, a first tag indicating a change above a first change threshold has occurred in the second media frame relative to the first media frame. The process can further include generating, using a machine learning model, a second tag indicating that media content of the second media frame is associated with a particular type of media content. The process can further include determining, based the first tag and the second tag, that the media content of the second media frame is associated with the particular type of media content.

Claims (67)

1. A method of processing media content, the method comprising:

detecting a switch of a device from being tuned to a first channel to being tuned to a second channel;

based on detecting the switch, obtaining a first media frame, a second media frame, and a third media frame associated with the first channel, the third media frame occurring after the first media frame and the second media frame;

generating, based on processing the first media frame and the second media frame using a first change detector, a first tag indicating a change above a first change threshold has occurred in the second media frame relative to the first media frame;

generating, based on processing the second media frame using a machine learning model that is separate from the first change detector, a second tag indicating that media content of the second media frame is associated with a particular type of media content;

determining, based on a combined value generated based on the first tag generated by the first change detector and the second tag generated by the machine learning model, that the media content of the second media frame is associated with the particular type of media content, wherein the device stays tuned to the second channel based the media content of the second media frame being associated with the particular type of media content; and

at least one of automatically tuning the device from the second channel to the first channel or outputting a notification based on determining that media content of the third media frame is not associated with the particular type of media content, the notification including an option to tune the device from the second channel to the first channel.

2. The method of claim 1 , further comprising:

generating, using the first change detector, a third tag indicating a change above the first change threshold has occurred in the third media frame;

generating, using the machine learning model, a fourth tag indicating a likelihood that media content of the third media frame is not associated with the particular type of media content; and

determining, based on the third tag and the fourth tag, that the media content of the third media frame is not associated with the particular type of media content.

3. The method of claim 1 , further comprising:

automatically tuning the device from the second channel to the first channel based on receiving input indicating a selection of the option to tune the device from the second channel to the first channel.

4. The method of claim 1 , further comprising:

segmenting the second media frame into a background region and one or more foreground regions;

comparing at least one of the background region and the one or more foreground regions of the second media frame to at least one of a background region and one or more foreground regions of the first media frame; and

determining, by the first change detector based on the comparing, that the change is above the first change threshold in the second media frame relative to the first media frame.

5. The method of claim 4 , further comprising:

determining, by the first change detector, that the background region and the one or more foreground regions of the second media frame have changed relative to the background region and the one or more foreground regions of the first media frame; and

determining, by the first change detector, that the change is above the first change threshold in the second media frame relative to the first media frame based on the background region and the one or more foreground regions of the second media frame having changed relative to the background region and the one or more foreground regions of the first media frame.

6. The method of claim 1 , further comprising:

comparing the second media frame to the first media frame; and

determining, by the first change detector based on the comparing, that the change is above the first change threshold in the second media frame relative to the first media frame.

7. The method of claim 6 , wherein comparing the second media frame to the first media frame includes comparing pixels of the second media frame to corresponding pixels of the first media frame.

8. The method of claim 6 , wherein comparing the second media frame to the first media frame includes comparing one or more statistical characteristics of a group of pixels of the first media frame to one or more statistical characteristics of a corresponding group of pixels of the second media frame.

9. The method of claim 6 , wherein comparing the second media frame to the first media frame includes comparing blocks of the first media frame to corresponding blocks of the second media frame.

10. The method of claim 6 , wherein comparing the second media frame to the first media frame includes comparing a color histogram of the first media frame to a color histogram of the second media frame.

11. The method of claim 1 , further comprising:

generating, using a second change detector, a third tag indicating a change above a second change threshold has occurred in the second media frame relative to the first media frame; and

determining, based on the first tag, the second tag, and the third tag, whether the media content of the second media frame is associated with the particular type of media content.

12. The method of claim 11 , further comprising:

generating, using a third change detector, a fourth tag indicating a change above a third change threshold has occurred in the second media frame relative to the first media frame; and

determining, based on the first tag, the second tag, the third tag, and the fourth tag, whether the media content of the second media frame is associated with the particular type of media content.

13. The method of claim 12 , wherein:

the first change detector determines change based on comparing at least one of a background region and one or more foreground regions of the second media frame to at least one of a background region and one or more foreground regions of the first media frame;

the second change detector determines change based on comparing the second media frame to the first media frame; and

the third change detector determines change based a comparison of audio associated with the second media frame with audio associated with the first media frame.

14. The method of claim 1 , wherein the machine learning model includes a first neural network and a second neural network that is different from the first neural network, the first neural network having a same configuration and a same set of parameters as the second neural network, the method further comprising:

generating a first output based on processing the first media frame using the first neural network;

generating a second output based processing the second media frame using the second neural network;

determining that the first output matches the second output; and

determining, based on determining that the first output matches the second output, that the media content of the second media frame is associated with the particular type of media content.

15. The method of claim 1 , wherein determining, based on the first tag and the second tag, that the media content of the second media frame is associated with the particular type of media content includes:

determining that a combined weight associated with the first tag and the second tag is greater than a weight threshold; and

determining, based on determining that the combined weight is greater than the weight threshold, that the media content of the second media frame is associated with the particular type of media content.

16. A system comprising:

a storage configured to store instructions; and

a processor configured to execute the instructions and cause the processor to:

detect a switch of a device from being tuned to a first channel to being tuned to a second channel;

based on detecting the switch, obtain a first media frame, a second media frame, and a third media frame associated with the first channel, the third media frame occurring after the first media frame and the second media frame,

generating, based on processing the first media frame and the second media frame use a first change detector, a first tag indicating a change above a first change threshold has occurred in the second media frame relative to the first media frame,

generate, based on processing the second media frame using a machine learning model that is separate from the first change detector, a second tag indicating that media content of the second media frame is associated with a particular type of media content,

determine, based on a combined value generated based on the first tag generated by the first change detector and the second tag generated by the machine learning model, that the media content of the second media frame is associated with the particular type of media content, wherein the device stays tuned to the second channel based the media content of the second media frame being associated with the particular type of media content; and

at least one of automatically tune the device from the second channel to the first channel or output a notification based on determining that media content of the third media frame is not associated with the particular type of media content, the notification including an option to tune the device from the second channel to the first channel.

17. The system of claim 16 , wherein the processor is configured to execute the instructions and cause the processor to:

generate, use the first change detector, a third tag indicating a change above the first change threshold has occurred in the third media frame;

generate, use the machine learning model, a fourth tag indicating a likelihood that media content of the third media frame is not associated with the particular type of media content; and

determine, based on the third tag and the fourth tag, that the media content of the third media frame is not associated with the particular type of media content.

18. The system of claim 16 , wherein the processor is configured to execute the instructions and cause the processor to:

segment the second media frame into a background region and one or more foreground regions;

compare at least one of the background region and the one or more foreground regions of the second media frame to at least one of a background region and one or more foreground regions of the first media frame; and

determine, by the first change detector based on the comparing, that the change is above the first change threshold in the second media frame relative to the first media frame.

19. The system of claim 16 , wherein the processor is configured to execute the instructions and cause the processor to:

compare the second media frame to the first media frame; and

determine, by the first change detector based on the comparing, that the change is above the first change threshold in the second media frame relative to the first media frame.

20. The system of claim 16 , wherein the processor is configured to execute the instructions and cause the processor to:

automatically tune the device from the second channel to the first channel based on receiving input indicating a selection of the option to tune the device from the second channel to the first channel.

Assignments (4)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 7, 2023
From: BENNETT-JAMES, JONATHAN
To: NAGRA MEDIA UK LTD
Reel/Frame 063261/0762 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 7, 2023
From: SINGH, BINEET KUMAR; KUMAR, NISHANT
To: NAGRAVISION INDIA PRIVATE LIMITED
Reel/Frame 063261/0875 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 7, 2023
From: NAGRA MEDIA UK LIMINTED
To: NAGRAVISION S.A.
Reel/Frame 063261/0923 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 7, 2023
From: NAGRAVISION INDIA PRIVATE LIMITED
To: NAGRAVISION S.A.
Reel/Frame 063262/0007 →
Continuity (2)
Provisional Application 63123259 · Dec 9, 2020
Related Publication 20220182430A1 · Jun 9, 2022