# US 10,503,980 B2 · System and method for automatic categorization of audio/video content

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 10,503,980 B2
- Document ID: US10503980B2
- Status: Granted
- Kind: B2
- Application number: 16/172,898
- Filed: 2018-10-29
- Published: 2019-05-02
- Granted: 2019-12-10
- Earliest claimed: 2017-10-31
- Assignees: Advanced Digital Broadcast S.A.
- Applicants: Advanced Digital Broadcast S.A.
- Inventors: Damian Siusta
- Web page: https://patentgrep.com/patent/US10503980B2

## CPC

- G06V 20/47: {Detecting features for summarising video content}
- G06V 20/48: {Matching video sequences}
- G06V 20/49: {Segmenting video sequences, i.e. computational techniques such as parsing or cutting the sequence, low-level clustering or determining units such as shots or scenes}
- H04N 21/44008: {involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream (arrangements characterised by components specially adapted for monitoring, identification or recognition of video in broadcast systems H04H60/59)}
- H04N 21/4542: {Blocking scenes or portions of the received content, e.g. censoring scenes}
- H04N 21/812: {involving advertisement data (advertising per se G06Q30/02)}
- H04N 21/8456: {by decomposing the content in the time domain, e.g. in time segments}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

Method for categorization of audio/video content, the method comprising the steps of: detecting a start and an end of at least one advertisements block present in said audio and/or video content; for each detected advertisements block: selectively collecting audio and/or video data frames, from said audio and/or video content, within specified time intervals prior to said detected advertisements block; for each time interval, computing a fingerprint; selectively collecting reference audio and/or video data frames, after said advertisements block, for comparison and computing a reference fingerprint for said reference audio and/or video data frames; comparing said reference fingerprint with at least one fingerprint collected prior to the start of said advertisements block in order to obtain a level of similarity between said fingerprints; based on the level of similarity, taking a decision whether to indicate a split point between content items present in said audio/video content.

## Claims

### Claim 1

Method for automatic categorization of content of an audio/video data stream, the method being characterized in that it comprises the steps of: accepting a request to execute said categorization of said content of an audio/video data stream; detecting a start and an end of at least one advertisements block present in said audio and/or video data stream; for each detected advertisements block: selectively collecting audio and/or video data frames, from said audio and/or video data stream, within specified time intervals prior to said detected advertisements block; for each time interval, for which audio and/or video data frames have been selectively collected, computing a fingerprint; selectively collecting reference audio and/or video data frames, after said advertisements block, for comparison and computing a reference fingerprint for said reference audio and/or video data frames; comparing said reference fingerprint with at least one fingerprint collected prior to the start of said advertisements block in order to obtain a level of similarity between said fingerprints; based on the level of similarity, taking a decision whether to indicate a split point between distinct content items present in said audio/video data stream.

### Claim 2 (depends on 1)

The method according to claim 1 wherein said selective collecting concerns collecting N of consecutive video data frames at defined timings.

### Claim 3 (depends on 2)

The method according to claim 2 wherein a type of said video data frame is taken into consideration during said selective collecting.

### Claim 4 (depends on 1)

The method according to claim 1 wherein said fingerprint is a data set, smaller than the referred data, that may be compared in order to determine a similarity level with another fingerprint.

### Claim 5 (depends on 1)

The method according to claim 1 wherein when determining said level of similarity there is executed one of: computing an average similarity level among all fingerprints pairs; selecting a subset of top results; setting a threshold for a top result.

### Claim 6 (depends on 1)

The method according to claim 1 wherein at least one of said time intervals falls after said end of the advertisements block.

### Claim 7 (depends on 1)

The method according to claim 1 wherein said an audio/video data stream is initially processed so that each advertisements section is removed in order to arrive at an output content.

### Claim 8 (depends on 7)

The method according to claim 7 wherein original timing information is preserved and it is known to the system to which time in the audio/video data stream the respective parts of the output content correspond.

### Claim 9 (depends on 1)

A non-transitory computer readable medium storing computer-executable instructions performing all the steps of the computer-implemented method according to claim 1 when executed on a computer.

### Claim 10 (depends on 1)

System for automatic categorization of content of an audio/video data stream, the system comprising: an audio and/or video data stream reception block; a memory configured to store temporary data as well as audio/video data stream; the system being characterized in that it further comprises: a controller configured to execute all steps of the method according to claim 1.