US 12,182,192 B1Granted

Content identification using fingerprinting

AssigneeAmazon Technologies, Inc.

InventorsNagaraj Mahajan, Ahmed Abdelal, Sumit Garg, Sai Kiran Venkata Subramanya Rupanagudi

Life of this patent

Estimating the term…

Loading patent…

Abstract

A system configured to perform content identification using fingerprinting to recognize known media content. The system may generate a reference database including reference fingerprints for each media content item to include in the content identification. In addition, the system may generate a hash table that associates individual frames of the reference fingerprints with identification information for corresponding media content items. When a device is playing media content, the system may perform content identification by generating query fingerprints representing the media content and comparing the query fingerprints to the reference database. For example, the system may match a query fingerprint to a reference fingerprint by identifying which of the reference fingerprints shares the most frames with the query fingerprint using the hash table. In addition, the system may use additional decision criteria to confirm a match, such as fine-grain matching or tracking successive fingerprints over time.

Claims21

Dependent claims

A computer-implemented method, the method comprising:

receiving first data corresponding to an audio fingerprint representation of audio data, the first data including first feature data corresponding to a first frame of the audio data and second feature data corresponding to a second frame of the audio data;
determining, using a hash function and the first feature data as a first hash key, a first hash value of a first plurality of hash values, the first hash value corresponding to the first feature data, the hash function mapping a second plurality of hash keys to the first plurality of hash values, wherein the first frame of the audio data only corresponds to the first hash value of the first plurality of hash values;
determining, using the hash function and the second feature data as a second hash key, a second hash value of the first plurality of hash values, the second hash value corresponding to the second feature data, wherein the second frame of the audio data only corresponds to the second hash value of the first plurality of hash values;
using a hash table to determine that the first hash value is associated with first identification data corresponding to a first media content item, the hash table including (i) first association data associating the first identification data and first values of the first plurality of hash values and (ii) second association data associating second identification data and second values of the first plurality of hash values, the second identification data corresponding to a second media content item;
determining, using the hash table, that the second hash value is associated with the first identification data and with the second identification data;
determining a first number of frames of the audio data that are associated with the first identification data;
determining a second number of frames of the audio data that are associated with the second identification data;
determining that the first number of frames is greater than the second number of frames;
and
in response to determining that the first number of frames is greater than the second number of frames, determining that the audio data represents the first media content item.

7 more elements

A computer-implemented method, the method comprising:

determining first data including first feature data representing a first segment of input data including a first frame of the input data and second feature data representing a second segment of the input data including a second frame of the input data;
determining, using a hash function and the first feature data as a first hash key, a first hash value corresponding to the first segment of the input data;
determining, using the hash function and the second feature data as a second hash key, a second hash value corresponding to the second segment of the input data;
determining, using a hash table, that the first hash value is associated with first identification data;
determining, using the hash table, that the second hash value is associated with the first identification data and second identification data;
determining a first number of hash keys of the first data that are associated with the first identification data;
determining a second number of hash keys of the first data that are associated with the second identification data;
determining that the first number of hash keys is greater than the second number of hash keys;
and
storing an association between the first data and the first identification data.

7 more elements

A system comprising:

at least one processor;
and
memory including instructions operable to be executed by the at least one processor to cause the system to:
determine first data including first feature data representing a first segment of input data including a first frame of the input data and second feature data representing a second segment of the input data including a second frame of the input data;
determine, using a hash function and the first feature data as a first hash key, a first hash value corresponding to the first segment of the input data;
determine, using the hash function and the second feature data as a second hash key, a second hash value corresponding to the second segment of the input data;
determine, using a hash table, that the first hash value is associated with first identification data;
determine, using the hash table, that the second hash value is associated with the first identification data and second identification data;
determine a first number of hash keys of the first data that are associated with the first identification data;
determine a second number of hash keys of the first data that are associated with the second identification data;
determine that the first number of hash keys is greater than the second number of hash keys;
and
store an association between the first data and the first identification data.

10 more elements

Dependencies are inferred from the claim text.

Facts

Application: 17/854,219Filed 30 Jun 2022
Granted: 31 Dec 2024
Term ends: Not estimatedThe source records don’t support an estimate.

Classification10 codes

G06F 16/41
Of multimedia data, e.g. slideshows comprising image and additional audio data: indexing; Data structures therefor; Storage structures
G06F 16/61
Of audio data: indexing; Data structures therefor; Storage structures
G06F 16/632
Querying: query formulation
G06V 20/46
In video content: extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames

Family

No related US applications are recorded in the indexed filing history.

Timeline

Citations

Open elsewhere

Building an agent or pulling many patents? Use the API or MCP server instead of scraping pages. Get an API key