Video Object Data Storage Using Neural Network Classification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current digital video systems lack the ability for users to interact with videos and have inadequate data storage techniques, limiting their effectiveness in digital advertising, entertainment, and communication.

Innovation Solution

A video processing system that includes end-user devices, video object storing devices, object classification devices using neural networks, and communication control devices to recognize, store, and manage video objects, allowing for better interaction and storage of video data, including object classification, metadata association, and efficient data storage methods.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If video data is stored in traditional formats, then storage capacity is maintained, but user interaction capability and data accessibility are limited

Engineering Contradiction:
Improveuser interaction capabilityVSAvoidstorage system complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments video data into distinct objects with individual metadata, allowing independent access and interaction with specific video elements. This segmentation enables users to interact with particular objects rather than viewing the entire video, thereby improving adaptability and user interaction capability while maintaining manageable storage through structured organization.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent adds a metadata dimension to traditional video storage by associating structured data (bounding boxes, timestamps, object types) with video objects. This dimensional enhancement enables new interaction modes and search capabilities without fundamentally changing the core storage mechanism, thus improving versatility without excessive complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If neural networks are used for object classification, then object recognition accuracy is improved, but processing time and computational resources increase

Engineering Contradiction:
Improveobject recognition accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs object classification and metadata generation as a preliminary action during video processing, so that when users interact with video objects, the recognition and classification are already complete. This eliminates real-time processing delays and allows the system to leverage pre-computed accurate classifications without adding processing time to user interactions.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If detailed metadata is stored for each video object, then data accessibility and interaction are improved, but storage requirements increase

Engineering Contradiction:
Improvedata accessibilityVSAvoidstorage requirements
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The patent creates structured metadata copies of video object information (bounding boxes, timestamps, object types) that can be efficiently stored and queried. Instead of storing redundant video data, the system maintains compact metadata representations that enable rapid access and interaction while minimizing storage requirements through efficient data encoding.

Inventive Principle:
Principle #26Copying

Data Source

PatentUS10217001B2Video object data storage and processing system
Publication Date: 2019.02.26 KICKVIEW CORP
  • US10217001B2 patent drawing
  • US10217001B2 patent drawing
  • US10217001B2 patent drawing

AI summary

A video object data storage and display system comprising a video object data selection and viewing portion and a video object data storage portion. The system comprises a video object having a: scale/size, pose/tilt, location, and frame/time. The system further comprises a database.