Embedding Secondary Object Data in Primary Video Streams

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Network-based video surveillance systems face increased data transfer and storage demands when multiple cameras capture the same scene from different angles, leading to bandwidth and storage inefficiencies, particularly when additional cameras are added without sufficient network resources.

Innovation Solution

A system that embeds secondary object data from a second camera into the primary video stream of a first camera, using a controller to determine object quality metrics and adjust camera views, embed object feature data, and process it for analytics, thereby reducing the need for additional video streams.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple cameras are positioned to provide multiple views of the same scene, then object recognition capability is improved, but data transfer and storage requirements increase

Engineering Contradiction:
Improveobject recognition capabilityVSAvoiddata transfer and storage requirements
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent extracts only the essential object data from the secondary camera's video stream and embeds it as metadata in the primary video stream. This allows the system to utilize multiple camera views for improved object recognition while avoiding the need to transmit and store complete duplicate video streams, thereby reducing data transfer and storage requirements.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent embeds secondary object data (metadata) within the primary video stream structure. The secondary camera's object information is nested as metadata tracks or embedded data within the primary video frame structure, allowing multiple data sources to be combined in a hierarchical manner without requiring separate transmission channels for each camera.

Inventive Principle:
Principle #7Nested doll (Nesting)

2Adaptability or versatility

If a second camera is added to an existing system, then viewing coverage is improved, but network bandwidth requirements increase

Engineering Contradiction:
Improveviewing coverageVSAvoidnetwork bandwidth requirements
Core Design Contradiction:
Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The system extracts only the necessary object detection and recognition data from the second camera rather than transmitting the complete video stream. This selective extraction of essential information allows the system to expand viewing coverage while minimizing the additional network bandwidth required.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the data format and representation parameters by converting secondary camera data from full video streams into compressed metadata formats with standardized schemas. This parameter transformation significantly reduces the data volume that needs to be transmitted over the network while preserving the essential viewing coverage information.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS11496671B2Surveillance video streams with embedded object data
Publication Date: 2022.11.08 SANDISK TECHNOLOGIES LLC
  • US11496671B2 patent drawing
  • US11496671B2 patent drawing
  • US11496671B2 patent drawing

AI summary

Systems and methods for surveillance video streams with embedded object data from another video camera are described. At least two video cameras are configured with fields of view to provide images of an object from alternative views. Video data for a primary video stream is received from one camera and secondary object data for the object from the other camera is embedded in the primary video stream. The primary video stream is sent to an analytics engine for processing the primary video and embedded secondary object data, such as performing facial recognition on a better image of a human face and/or feature vectors therefrom that are embedded in the primary video stream.