IP Library › Granted Patent US 12,501,104
Granted Patent B2
US 12,501,104 · App. 18/103,043 · Granted Dec 16, 2025

Systems and methods for leveraging machine learning for improving and enhancing DVR services

Inventors: Johan Kölhi (Vaxholm, SE); Anthony Friede (Upplands Väsby, SE); Reda Harb (Issaquah, WA)
Assignee: Adeia Guides Inc.
H04N21/4334G06V10/764H04N21/44004H04N21/47217G06V2201/07
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 12,501,104
App. No.
18/103,043
Granted
Dec 16, 2025
Kind
B2
Abstract

A media stream is accessed, and a classification model is selected for identifying objects in the media stream. Using the classification model, it is determined whether a content item of interest to the user is currently being received in the media stream. If the content item is currently being received in the media stream, at least a portion of the content item is recorded from the media stream. If the content item is not currently being received, then recording is paused or delayed until it is determined that the content item is being received.

Claims (62)

1 . A method for enhancing digital video recording services, the method comprising:

accessing, at a client device, a media stream based on a start time of a scheduled recording of a content item;

monitoring the media stream while recording the content item beginning at the scheduled start time;

selecting a classification model;

based at least in part on monitoring the media stream while recording the content item beginning at the scheduled start time, identifying, using the selected classification model, objects in the media stream while recording the content item;

determining an identified object of a plurality of identified objects is of interest to a user;

in response to determining the identified object of interest to the user of the plurality of identified objects is not currently being received in the media stream during the recording of the content item:

pausing recording of the content item from the media stream;

continuing to monitor the media stream while the recording of the content item from the media stream is paused; and

in response to determining, based on the continued monitoring, that the identified object of interest to the user of the plurality of identified objects is again being received in the media stream, resuming recording of the content item from the media stream.

2 . The method of claim 1 , further comprising:

determining, using the classification model, whether a current portion of the content item is of interest to the user;

in response to determining that the current portion of the content item is of interest to the user, recording the current portion of the content item; and

in response to determining that the current portion of the content item is not of interest to the user, pausing recording of the content item.

3 . The method of claim 1 , further comprising:

buffering a portion of the media stream; and

in response to determining that the content item is currently being received in the media stream, recording at least a portion of the buffered portion of the media stream corresponding to content received while the determining occurred.

4 . The method of claim 1 , further comprising determining whether the content item is of interest to the user.

5 . The method of claim 4 , wherein determining whether the content item is of interest to the user further comprises:

identifying, based on a content consumption history of the user, an object;

determining, using the classification model, whether the object is present in the media stream; and

in response to determining that the object is present in the media stream, identifying a content item currently being received in the media stream as a content item of interest to the user.

6 . The method of claim 1 , further comprising:

receiving, during playback of the recorded content item, a request to navigate through the content item, wherein the request indicates an object, entity, or action;

selecting, based on the object, entity, or action indicated in the request, the classification model;

identifying, using the selected classification model, a playback position within the recorded content item that matches the request; and

playing back the recorded content item beginning at the identified playback position.

7 . The method of claim 1 , further comprising:

identifying, based on the classification model and a content consumption history of the user, playback points of interest to the user within the content item; and

generating a bookmark for each respective identified playback point of interest.

8 . A system for enhancing digital video recording services, the system comprising:

input/output circuitry; and

control circuitry configured to:

access, at a client device, a media stream based on a start time of a scheduled recording of a content item;

monitor the media stream while recording the content item beginning at the scheduled start time;

select a classification model;

based at least in part on monitoring the media stream while recording the content item beginning at the scheduled start time, identify, using the selected classification model, objects in the media stream while recording the content item;

determine an identified object of a plurality of identified objects is of interest to a user;

in response to determining the identified object of interest to the user of the plurality of identified objects is not currently being received in the media stream during the recording of the content item:

pause recording of the content item from the media stream;

continue to monitor the media stream while the recording of the content item from the media stream is paused; and

in response to determining, based on the continued monitoring, that the identified object of interest to the user of the plurality of identified objects is again being received in the media stream, resume recording of the content item from the media stream.

9 . The system of claim 8 , wherein the control circuitry is further configured to:

determine, using the classification model, whether a current portion of the content item is of interest to the user;

in response to determining that the current portion of the content item is of interest to the user, record the current portion of the content item; and

in response to determining that the current portion of the content item is not of interest to the user, pause recording of the content item.

10 . The system of claim 8 , wherein the control circuitry is further configured to:

buffer a portion of the media stream; and

in response to determining that the content item is currently being received in the media stream, record at least a portion of the buffered portion of the media stream corresponding to content received while the determining occurred.

11 . The system of claim 8 , wherein the control circuitry is further configured to determine whether the content item is of interest to the user.

12 . The system of claim 11 , wherein the control circuitry configured to determine whether the content item is of interest to the user is further configured to:

identify, based on a content consumption history of the user, an object;

determine, using the classification model, whether the object is present in the media stream; and

in response to determining that the object is present in the media stream, identify a content item currently being received in the media stream as a content item of interest to the user.

13 . The system of claim 8 , wherein the control circuitry is further configured to:

receive, during playback of the recorded content item, a request to navigate through the content item, wherein the request indicates an object, entity, or action;

select, based on the object, entity, or action indicated in the request, the classification model;

identify, using the selected classification model, a playback position within the recorded content item that matches the request; and

play back the recorded content item beginning at the identified playback position.

14 . The system of claim 8 , wherein the control circuitry is further configured to:

identify, based on the classification model and a content consumption history of the user, playback points of interest to the user within the content item; and

generate a bookmark for each respective identified playback point of interest.

Assignments (2)
SECURITY INTEREST Recorded May 3, 2023
From: ADEIA GUIDES INC.; ADEIA IMAGING LLC; ADEIA MEDIA HOLDINGS LLC; ADEIA MEDIA SOLUTIONS INC.; ADEIA SEMICONDUCTOR ADVANCED TECHNOLOGIES INC.; ADEIA SEMICONDUCTOR BONDING TECHNOLOGIES INC.; ADEIA SEMICONDUCTOR INC.; ADEIA SEMICONDUCTOR SOLUTIONS LLC; ADEIA SEMICONDUCTOR TECHNOLOGIES LLC; ADEIA SOLUTIONS LLC
To: BANK OF AMERICA, N.A., AS COLLATERAL AGENT
Reel/Frame 063529/0272 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 1, 2023
From: KOLHI, JOHAN; FRIEDE, ANTHONY; HARB, REDA
To: ADEIA GUIDES INC.
Reel/Frame 063496/0834 →
Continuity (1)
Related Publication 20240259629A1 · Aug 1, 2024
References Cited (8)
US 9138652B1 · Thompson · 2015 [cited by examiner]
US 11272248B2 · Neumeier et al. · 2022 [cited by applicant]
US 20090018920A1 · Lerman et al. · 2009 [cited by applicant]
US 20090080857A1 · St. John-Larkin · 2009 [cited by examiner]
US 20110251896A1 · Impollonia et al. · 2011 [cited by applicant]
US 20160203843A1 · Beals · 2016 [cited by examiner]
US 20200413113A1 · Tofighbakhsh · 2020 [cited by examiner]
US 20220101452A1 · Systrom et al. · 2022 [cited by applicant]