Video Object Extraction for Efficient Review

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image display technologies face challenges in efficiently highlighting and reviewing objects of interest, such as license plates or faces, within complex video data, requiring significant user effort due to their small size and integration within images.

Innovation Solution

A method and apparatus for detecting predefined types of objects within video frames, extracting them, and displaying them in sub-regions of a separate display area, allowing for sequential arrangement and prioritization, with user-selectable filtering and playback options to enhance visibility and user interaction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If video data is displayed in display regions for user review, then multiple video sources can be monitored, but objects of interest become difficult to identify due to their small size and integration within complex images

Engineering Contradiction:
Improvenumber of video streams monitoredVSAvoidease of reviewing objects of interest
Core Design Contradiction:
Quantity of substanceVSEase of operation

Solution Approach 1:

The patent extracts objects of interest (such as faces, license plates, or other predefined types) from the video frames and displays them separately in a dedicated display region. This extraction process isolates the objects from the complex video background, making them easily identifiable and reviewable without requiring users to search through entire video frames.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The display area is segmented into multiple regions: a first display region for showing video frames and a second display region for showing extracted objects. This spatial segmentation allows simultaneous viewing of both the context (video frames) and the objects of interest, resolving the contradiction between monitoring multiple streams and easily identifying specific objects.

Inventive Principle:
Principle #1Segmentation

2Loss of information

If objects are displayed within the original video frames, then spatial context is preserved, but user effort increases significantly due to the small size and complexity of images

Engineering Contradiction:
Improvespatial context preservationVSAvoiduser review time
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The patent introduces an intermediary processing step that detects and extracts objects of interest from video frames. This intermediary process creates a bridge between the original video data and the user's review process, preserving spatial context information while dramatically reducing the time users need to spend searching for objects, as extracted objects are presented in a dedicated display region.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Loss of information

If all video data is displayed simultaneously, then complete information is available, but objects of interest are lost among the complexity of the images

Engineering Contradiction:
Improveinformation completenessVSAvoiddifficulty of identifying objects of interest
Core Design Contradiction:
Loss of informationVSDifficulty of detecting and measuring

Solution Approach 1:

The system extracts specific objects of interest from the complete video data and presents them separately. This extraction maintains information completeness by preserving all original video data in the first display region while simultaneously making objects of interest easily detectable through separate display in the second region, thus resolving the difficulty of identification.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP3573327B1Method and device for displaying target object
Publication Date: 2024.04.03 HANGZHOU HIKVISION DIGITAL TECHNOLOGY CO LTD
  • EP3573327B1 patent drawingFigure 1~2
  • EP3573327B1 patent drawingFigure 3~5(d)
  • EP3573327B1 patent drawingFigure 6

AI summary

Embodiments of the present application provide a method and apparatus for displaying objects. In the method, video frames that contain objects of predefined types are detected in pieces of video data to be displayed in a first display region. If such video frames are detected, the objects of the predefined types are extracted from the detected video frames. The extracted objects are then displayed in sub-regions of a second display region that is different from the first display region, with each object being displayed in one of the sub-regions. In this way, the objects that a user is interested in can be extracted from a video, and displayed in a way to be viewed with ease.