Method and system for map target tracking

A method of tracking a map target through a map target tracking application executed by at least one processor of a terminal, the method comprising:

• acquiring a basic image obtained by photographing a 3D space;
• acquiring a plurality of sub-images obtained by dividing the acquired basic image for respective sub-spaces in the 3D space;
• creating a plurality of sub-maps based on the plurality of acquired sub-images;
• determining at least one main key frame for each of the plurality of created sub-maps;
• creating a 3D main map by combining the plurality of sub-maps for which the at least one main key frame is determined;
• and
• tracking current posture information in the 3D space based on the created 3D main map,
• wherein the determining of at least one main key frame includes at least one of:
• determining at least some key frames at a predetermined interval among a plurality of key frames for each sub-map as the main key frames of the corresponding sub-map;
• and
• calculating a similarity between the plurality of key frames for each sub-map and determining at least some key frames having the calculated similarity less than a predetermined criterion as the main key frames of the corresponding sub-map.

A method of tracking a map target through a map target tracking application executed by at least one processor of a terminal, the method comprising:

• acquiring a basic image obtained by photographing a 3D space;
• acquiring a plurality of sub-images obtained by dividing the acquired basic image for respective sub-spaces in the 3D space;
• creating a plurality of sub-maps based on the plurality of acquired sub-images;
• determining at least one main key frame for each of the plurality of created sub-maps;
• creating a 3D main map by combining the plurality of sub-maps for which the at least one main key frame is determined;
• and
• tracking current posture information in the 3D space based on the created 3D main map,
• wherein the tracking of current posture information in the 3D space based on the created 3D main map includes:
• acquiring a position detection image obtained by separately photographing the 3D space;
• calculating a similarity for each sub-map by comparing the acquired position detection image with the 3D main map;
• determining a sub-map having the highest similarity among the calculated similarities for the respective sub-maps as a representative sub-map;
• and
• detecting the current posture information that provides position and orientation information of a camera at each time when capturing each frame in the position detection image by comparing the determined representative sub-map with the position detection image.

A system for tracking a map target, the system comprising:

• at least one display for outputting a 3D main map;
• at least one memory;
• at least one processor;
• and
• at least one application stored in the memory and executed by the processor to perform the map target tracking,
• wherein the at least one application is operated to:
• acquire a basic image obtained by photographing a 3D space;
• acquire a plurality of sub-images obtained by dividing the acquired basic image for respective sub-spaces in the 3D space;
• create a plurality of sub-maps based on the plurality of acquired sub-images;
• determine at least one main key frame for each of the plurality of created sub-maps;
• create the 3D main map by combining the plurality of sub-maps for which the at least one main key frame is determined;
• track current posture information in the 3D space based on the created 3D main map;
• and
• acquire camera posture information that provides position and orientation information of a camera when capturing each frame in the basic image;
• extract at least one key frame for each sub-image;
• detect a plurality of feature points for each extracted key frame and a descriptor for each feature point;
• and
• create the plurality of sub-maps by mapping the detected feature points and the detected descriptor for each feature point to 3D space coordinates based on the camera posture information.