Method and system for determining the dimensions of an object shown in a multimedia content item

• US 10,607,355 B2
• Filed: 01/29/2015
• Issued: 03/31/2020
• Est. Priority Date: 10/26/2005
• Status: Active Grant

• Current Assignee: Cortica Ltd.
• Original Assignee: Cortica Ltd.
• Inventors: Raichelgauz, Igal, Odinaev, Karina, Zeevi, Yehoshua Y.
• Primary Examiner(s): Chang, Li Wu
• Application Number:
  US14/608,880
• Publication Number:
  US 20150139569A1
• Time in Patent Office: 1,888 Days
• Field of Search: None
• CPC Class Codes: G06F 16/434 using image data, e.g. images, photos, pictures taken by a user
• CPC Class Codes: G06F 16/7837 using objects detected or recognised in the video content
• CPC Class Codes: G06F 16/957 Browsing optimisation, e.g. caching or content distillation
• CPC Class Codes: G06T 1/0021 Image watermarking
• CPC Class Codes: G06T 7/60 Analysis of geometric attributes
• CPC Class Codes: G09B 19/0092 Nutrition
• CPC Class Codes: H04H 2201/90 characterised by the use of signatures
• CPC Class Codes: H04H 60/48 for recognising items expressed in broadcast information
• CPC Class Codes: H04H 60/59 of video recognising characters or patterns in general G06F18/00, G06V20/00
• CPC Class Codes: H04N 21/25891 being end-user preferences retrieval of video data in a video database based on user preferences G06F16/739; arrangements for recognizing users' preferences H04H60/46; user profiles in network data switching protocols H04L67/306; processing of user preferences or user profiles in wireless networks H04W8/18
• CPC Class Codes: H04N 21/2668 Creating a channel for a dedicated end-user group, e.g. insertion of targeted commercials based on end-user profiles information retrieval from the Internet by querying with filtering and personalisation G06F16/9535; arrangements for replacing or switching information during the broadcast H04H20/10; push services over packet-switching network H04L12/1859; adaptation of message content in packet-switching networks H04L51/063
• CPC Class Codes: H04N 21/466 Learning process for intelligent management, e.g. learning user preferences for recommending movies services using the results of monitoring in broadcast systems H04H60/61
• CPC Class Codes: H04N 21/8106 involving special audio data, e.g. different tracks for different languages
• CPC Class Codes: H04N 7/17318 Direct or substantially direct transmission and handling of requests

Abstract

A method and system for determining at least a size dimension of objects shown in multimedia content items are presented. The method includes receiving an input multimedia content item; identifying objects shown in the multimedia content item; generating at least a first signature for at least a first object of the plurality of objects and at least a second signature for at least a second object; identifying at least one concept that matches the at least a first object; determining an actual size of the first object respective of the match to the at least one concept; determining a size scale between the first object and the second object using the at least a first signature and the at least a second signature; and determining the at least size dimension of the second object respective of the size scale and the actual size of the first object.

Claims

1. A method for determining at least a size dimension of objects shown in multimedia content items, comprising:

receiving an input multimedia content item;

identifying a plurality of objects shown in the input multimedia content item;

generating at least one first signature for at least one first object of the plurality of objects and at least one second signature for at least one second object of the plurality of objects;

identifying, using the at least one first signature, at least one matching concept, wherein each matching concept of the at least one matching concept includes at least one signature that matches the at least one first signature;

wherein each matching concept comprises a cluster of signatures of multiple multimedia content items and metadata related to the multiple multimedia content items;

wherein each matching concept undergoes a cluster reduction process that comprises reducing a number of signatures of the matching cluster while remaining a minimal number of signatures that still identify all of the multiple multimedia content items;

wherein for each matching cluster the cluster reduction process comprises matching signatures of each of the multimedia content items associated with the matching cluster, with a reduced cluster having one or more signatures removed there from to provide results; and

determining whether to remove the one or more signatures from the concept based on the results;

determining an actual size of the at least one first object respective of the at least one matching concept, wherein the actual size of the at least one first object is determined respective of an actual size of the at least one matching concept maintained in a data warehouse;

determining a size scale between the at least one first object and the at least one second object of the plurality of objects using the at least one first signature and the at least one second signature; and

determining the at least size dimension of the at least one second object of the plurality of objects respective of the size scale and the actual size of the first object.

2. The method of claim 1, further comprising:

storing the determined size dimension of the at least one second object in the data warehouse.

3. The method of claim 1, wherein identifying the at least one matching concept further comprises:

querying, a deep content classification (DCC) system, using the at least one first signature to find the at least one matching concept.

4. The method of claim 1, wherein the size scale between the at least one first object and the at least one second object of the plurality of objects is determined respective of at least one of:

a ratio between the at least one first signature and the at least one second signature, and the distance of the at least one first object and the at least one second object from a reference point.

5. The method of claim 1, wherein the at least one first signature for at least one first object represents a response of multiple leaky integrated-to-threshold unit nodes to the at least one first object.

6. The method of claim 1, wherein the received multimedia content item is any one of:

an image, a graphic, a video frame, and a photograph.

7. A non-transitory computer readable medium having stored thereon instructions for causing one or more processing units to execute the method according to claim 1.

15. The method of claim 1, wherein the at least one first signature for at least one first object is generated by a plurality of at least partially statistically independent computational cores that comprise multiple leaky integrated-to-threshold unit nodes.

16. The method of claim 15, wherein the plurality of at least partially statistically independent computational cores are implemented in one or more integrated circuits.

8. A system for determining at least a size dimension of objects shown in a multimedia content item containing a plurality of objects, comprising:

an interface to a network for receiving an input multimedia content item;

a processing unit; and

a memory connected to the processing unit and configured to contain a plurality of instructions that when executed by the processor configure the system to;

identify a plurality of objects shown in the input multimedia content item;

identify, using the at least one first signature, at least one matching concept;

wherein each matching concept of the at least one matching concept includes at least one signature that matches the at least one first signature;

wherein each matching concept comprises a cluster of signatures of multiple multimedia content items and metadata related to the multiple multimedia content items;

wherein each matching concept undergoes a cluster reduction process that comprises reducing a number of signatures of the matching cluster while remaining a minimal number of signatures that still identify all of the multiple multimedia content items;

wherein for each matching cluster the cluster reduction process comprises matching signatures of each of the multimedia content items associated with the matching cluster, with a reduced cluster having one or more signatures removed there from to provide results; and

determining whether to remove the one or more signatures from the concept based on the results;

determine an actual size of the at least one first object respective of the at least one matching concept, wherein the actual size of the at least one first object is determined respective of an actual size of the at least one matching concept maintained in a data warehouse;

determine a size scale between the at least one first object and at least one second object of the plurality of objects using the at least one first signature and at least one second signature; and

determine the at least size dimension of the at least one second object of the plurality of objects respective of the size scale and the actual size of the at least one first object.

9. The system of claim 8, wherein the plurality of instructions further configure the system to:

store the determined size dimension of the at least one second object in the data warehouse.

10. The system of claim 8, further comprising:

a deep content classification (DCC) system for querying the at least one first signature to find the at least one matching concept, wherein the DCC is communicatively connected to the system.

11. The system of claim 8, wherein the signatures of each object are generated by a signature generator system (SGS), wherein the SGS is communicatively connected to the system.

12. The system of claim 11, wherein the least one first signature for at least one first object represents a response of multiple leaky integrated-to-threshold unit nodes to the at least one first object.

13. The system of claim 8, wherein the size scale between the at least one first object and the at least one second object of the plurality of objects is determined respective of at least one of:

a ratio between the at least one first signature and the at least one second signature, and the distance of the at least one first object and the at least one second object from a reference point.

14. The system of claim 8, wherein the received multimedia content item is any one of:

an image, a graphic, a video frame, and a photograph.

17. The system of claim 8, wherein the at least one first signature for at least one first object is generated by a plurality of at least partially statistically independent computational cores of the SGS that comprise multiple leaky integrated-to-threshold unit nodes.

18. The system of claim 17, wherein the plurality of at least partially statistically independent computational cores are implemented in one or more integrated circuits.