Surveillance Video 3D Background Integration via Grid Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for three-dimensional presentation of surveillance videos fail to deeply integrate the video with the background model, resulting in a non-intuitive representation, as each frame is presented separately on a pre-established three-dimensional background.

Innovation Solution

A method and apparatus that acquire a surveillance video and a preconfigured three-dimensional background model with video grid information, dividing and adjusting the video image into corresponding image areas based on the grid information to fit the three-dimensional background model, allowing for deep integration and enhanced presentation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the surveillance video is presented as a pop-up window on the three-dimensional background model, then the three-dimensional presentation effect is achieved, but the video cannot be deeply integrated with the background model

Engineering Contradiction:
Improveintegration depthVSAvoidpresentation complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The video image is divided into multiple video image areas according to video grid information, and each area is mapped to corresponding background model areas. This segmentation allows the video to be deeply integrated with the three-dimensional background model while maintaining manageable complexity through systematic processing of divided regions.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from two-dimensional video presentation to three-dimensional integration by mapping video image areas onto the three-dimensional background model. This dimensional transformation enables deep integration of video content with the spatial background, creating an immersive three-dimensional presentation effect.

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

2Ease of operation

If each frame of the surveillance video is presented in its entirety on the three-dimensional background model, then the complete video content is displayed, but the presentation is not intuitive

Engineering Contradiction:
ImproveintuitivenessVSAvoidspatial relationship information
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The video image is divided into multiple video image areas that correspond to different regions of the three-dimensional background model. This segmentation enables intuitive spatial presentation by mapping video content to appropriate background regions, preserving spatial relationships while improving visual intuitiveness.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different video image areas are mapped to different areas of the background model based on their spatial characteristics. This local quality approach ensures that each video region is presented in the most appropriate location on the three-dimensional model, enhancing overall intuitiveness while maintaining complete video information.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3448020B1Method and device for three-dimensional presentation of surveillance video
Publication Date: 2022.04.06 HANGZHOU HIKVISION DIGITAL TECHNOLOGY CO LTD
  • EP3448020B1 patent drawingFigure 1~2
  • EP3448020B1 patent drawingFigure 3~5

AI summary

Embodiments of the present application provide a method and apparatus for three-dimensional presentation of a surveillance video, applicable to an electronic device. The method includes: acquiring a surveillance video, a preconfigured three-dimensional background model and video grid information corresponding to a scene in which the surveillance video is captured; wherein the video grid information includes sizes of image areas in a video image of the surveillance video, and area coordinates of the image areas in the three-dimensional background model; dividing, according to the video grid information, the video image of the surveillance video into a plurality of image areas of corresponding sizes, and adjusting the image areas based on the area coordinates of the image areas in the three-dimensional background model, which are included in the video grid information; and superimposing the adjusted image areas on areas with corresponding coordinates in the three-dimensional background model. The surveillance video can be deeply integrated with the three-dimensional background model. A better three-dimensional presentation of the surveillance video is provided.