Visual Fingerprint Video Search Catalog

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video search methods rely on textual metadata and Boolean operators, which are ineffective in accurately representing and retrieving video or audio content, leading to poor performance in identifying and retrieving video content.

Innovation Solution

A video search and identification system that uses a catalog to represent relationships between video, data, and objects, enabling querying and search based on visual representations, including video nodes, metadata nodes, and card nodes, with visual fingerprints generated from frames to match against a base set of signatures, and confidence measures to adjust query processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If visual-based fingerprint matching is used, then search accuracy is improved, but device complexity increases

Engineering Contradiction:
Improvesearch accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The video frame is divided into multiple cells to generate visual fingerprints. Each cell's brightness is independently analyzed to create bit representations, allowing complex visual data to be segmented into manageable units for processing and comparison.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Visual fingerprints are created as simplified representations (copies) of the actual video frames. These fingerprints capture essential visual characteristics through bit patterns and confidence measures, enabling search operations without requiring processing of the full original video data.

Inventive Principle:
Principle #26Copying

2Measurement precision

If visual fingerprints and confidence measures are processed, then retrieval precision is improved, but processing time increases

Engineering Contradiction:
Improveretrieval precisionVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

Different cells in the visual fingerprint are assigned confidence measures based on their local quality and reliability. This allows the system to weight certain features more heavily than others during matching, improving precision while managing processing requirements by focusing computational effort where it matters most.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system transforms continuous visual data into discrete bit representations with associated confidence parameters. This parameter transformation enables efficient comparison operations while maintaining the essential information needed for precise retrieval.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If visual-based matching is implemented, then search effectiveness is improved, but difficulty of detecting and measuring increases

Engineering Contradiction:
Improvesearch effectivenessVSAvoidmeasurement difficulty
Core Design Contradiction:
ProductivityVSDifficulty of detecting and measuring

Solution Approach 1:

The system replaces complex visual analysis mechanisms with a simplified bit-based representation system. By converting visual frame data into binary fingerprints with confidence measures, the complexity of visual comparison is substituted with straightforward bit-matching operations that are easier to implement and measure.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS8719884B2Video identification and search
Publication Date: 2014.05.06 MICROSOFT TECHNOLOGY LICENSING LLC
  • US8719884B2 patent drawing
  • US8719884B2 patent drawing
  • US8719884B2 patent drawing

AI summary

Systems and methods for identifying and searching video are disclosed. A video search and identification system includes a catalog representing relationships between video, data and/or objects to enable querying and search based on visual representations of video as well as data or other information associated with the video. In one example, the catalog includes video nodes, metadata nodes and card nodes, although additional or fewer node types may be used. A visual-based video identification system is provided to identify content in video sources. An unidentified video source is accessed and visual fingerprints of one or more frames are generated as query signatures for matching against a base set of known signatures. Confidence measures are generated at the bit level to assist in query signature processing.