AR Content Identification Using 3D Models and Occlusion Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The recognition rate for buildings in an external environment decreases with increasing distance and interference from other objects, leading to inaccurate location and pose determination of electronic devices, which affects the accuracy of augmented reality content provision.
Innovation Solution
An electronic device processes images using a camera to obtain information on visual objects, transmits this information to a server, receives a 3D model and reference image, removes interfering objects, identifies feature points, and determines the pose of a virtual camera to accurately superimpose AR content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If building recognition is performed in the external environment, then AR content can be provided, but the recognition rate decreases with increasing distance and interference from other objects
Solution Approach 1:
The patent extracts and removes interfering objects (cars, persons, trees, etc.) from the captured image to isolate the building structure. This is achieved by comparing the captured image with reference images and identifying visual objects that are not part of the building, then removing them to improve recognition accuracy.
Solution Approach 2:
The patent performs preliminary actions by capturing multiple reference images of the building from different angles and positions before the actual AR content provision. These reference images are stored and used for comparison during the recognition process, enabling more accurate building identification even at distances or with partial occlusions.
2Area of stationary object
If building recognition is performed at a distance, then the field of view is expanded, but the recognition rate for buildings decreases
Solution Approach 1:
The patent makes the building recognition system universal by enabling it to recognize buildings at various distances and from multiple angles. By storing multiple reference images captured from different positions and using image comparison techniques, the system can identify buildings whether they are close or far away, maintaining recognition accuracy across different viewing conditions.
3Reliability
If multiple visual objects are present in the image, then the environment is more accurately captured, but the feature point identification becomes more difficult
Solution Approach 1:
The patent extracts and removes non-building visual objects (cars, persons, trees, etc.) from the captured image by comparing it with reference images. This isolation of the building structure simplifies the subsequent feature point identification process, making it easier to detect and measure building-specific features even when the original scene contained multiple distracting objects.
Data Source
AI summary
An electronic device includes a camera, a communication circuit, memory storing one or more computer programs, and at least one processor. The one or more computer programs include computer-executable instructions that, when executed by the one or more processors, cause the electronic device to obtain information on a second image comprising a plurality of visual objects and changed from a first image, transmit, to a server, the information on the second image, receive information on a three-dimensional (3D) model for a space comprising the plurality of external objects and information on a reference image, obtain a third image by removing, from the second image, at least one visual object among the plurality of visual objects, identify at least one feature point based on a comparison between the reference image and the third image, identify a pose of a virtual camera, and identify content superimposed on the first image.