Interaction system between infrastructure and multiple moving objects through sensor fusion based intersection 3D space creation and interaction method using the same

Patent metadata

Field	Value
Publication number	US-12482120-B2
Application number	US-202318367590-A
Country	US
Kind code	B2
Filing date	Sep 13, 2023
Priority date	Sep 14, 2022
Publication date	Nov 25, 2025
Grant date	Nov 25, 2025

Abstract

The interaction system includes: an infrastructure image sensor-based 3D cloud restoration module configured to create a 3D cloud point using a multi-view image received from an infrastructure sensor; a real-time 3D map creation and learning update module configured to generate subset data composed of point clouds into a 3D map using 3D point information received from the infrastructure image sensor-based 3D cloud restoration module; an object estimation module in a 3D space configured to recognize and track an object existing in a 3D space based on the 3D map received from the real-time 3D map creation and learning update module; and a dangerous situation recognition and interaction module in space configured to receive information on an object existing in the 3D space from the object estimation module in the 3D space, recognize a dangerous situation, and provide dangerous situation related information.

First claim

What is claimed is:

1 . An interaction system between infrastructure and multiple moving objects through sensor fusion based intersection 3D space creation, the interaction system comprising:
an infrastructure image sensor-based 3D cloud restoration module configured to create a 3D cloud point using a multi-view image received from an infrastructure sensor; a real-time 3D map creation and learning update module configured to generate subset data composed of point clouds into a 3D map using 3D point information received from the infrastructure image sensor-based 3D cloud restoration module; an object estimation module in a 3D space configured to recognize and track an object existing in a 3D space based on the 3D map received from the real-time 3D map creation and learning update module; and a dangerous situation recognition and interaction module in space configured to receive information on an object existing in the 3D space from the object estimation module in a 3D space, recognize a dangerous situation, and provide dangerous situation related information, wherein the real-time 3D map creation and learning update module restores the 3D space by occlusion between objects using sensor data acquired and provided from various angles and locations to generate the 3D map, wherein the real-time 3D map creation and learning update module generates the 3D map including sensor information existing in regional and global locations and extracts a correlation data between 3D map subsets, wherein the real-time 3D map creation and learning update module includes: a first layer configured to extract features by matching a subset of 3D points with an image, regenerate multiple images input at the same time as multi-view, and extract feature information on a line; a second layer configured to generate a multi-view warped image using the image received from the first layer and extract the features; a third layer configured to calculate viewpoint information using GPS information and sensor location information and provide building and road property information included in a 3D navigation or a web map; a fourth layer configured to integrate multi-view 3D point cloud features, feature information on the line, and property information on the building and road; and a fifth layer configured to train 3D point cloud information and generate a subset 3D map.

2 . The interaction system of claim 1 , wherein the infrastructure image sensor-based 3D cloud restoration module estimates a depth image of the multi-view image, and generates and manages static 3D information and dynamic 3D information.

3 . The interaction system of claim 1 , wherein the object estimation module in 3D space sets a 3D space range for recognizing the correlation and the dangerous situation between the moving object and a geographical feature existing in the space and for interaction.

4 . The interaction system of claim 3 , wherein the object estimation module in 3D space recognizes and tracks an object using the integrated 3D point information received from the real-time 3D map creation and learning update module, recognizes a type, a shape, and a state of the object, and confirms and manages the object existing in the restored 3D space based on a recognition result.

5 . The interaction system of claim 1 , wherein the dangerous situation recognition and interaction module in space recognizes a dangerous situation by using moving object information, road traffic related infrastructure information, and pedestrian information existing in the 3D space.

6 . The interaction system of claim 5 , wherein the dangerous situation recognition and interaction module in space trains information on the dangerous situation and response related information.

Assignees

• Electronics & Telecommunications Res Inst

Inventors

• Kim Jinwoo
• Kim Ki Tae
• Park Myungwook
• An Kyoung Hwan
• Choi Jeong Dan

Classifications

• G06V20/653

  by matching three-dimensional models, e.g. conformal mapping of Riemann surfaces · CPC title

• G06T2210/61

  Scene description · CPC title

• G06T17/00

  Three-dimensional [3D] modelling for computer graphics · CPC title

• G06T2207/30261

  Obstacle · CPC title

• G06T3/18

  Image warping, e.g. rearranging pixels individually · CPC title