Method of generating three-dimensional model data of object

1. A method, performed by a computing device, of generating three-dimensional model data of an object, the method comprising:

generating a first image by photographing an object positioned on a board including a marker with a camera at a first time point, and storing first view data of the camera at this time, the first view data of the camera being a matrix for converting a coordinate value on a world coordinate system onto a first camera coordinate system defined by a pose of the camera at the first time point;

generating first relative pose data of the object, based on the marker included in the first image, the first relative pose data of the object being a matrix for converting a coordinate value on an object coordinate system defined by the marker onto the first camera coordinate system;

estimating pose data of the object, based on the first view data of the camera and first relative pose data of the object, the pose data of the object being a matrix for converting a coordinate value on the object coordinate system onto the world coordinate system;

generating a second image by photographing the object with the camera at a second time point in which the marker is not in view of the camera and thus the marker is not included in the second image as photographed, and generating second view data of the camera at this time, wherein the position of the camera at the second time point is different from the position of the camera at the first time point;

estimating second relative pose data of the object, based on the pose data of the object and the second view data of the camera; and

generating three-dimension model data of the object based on the first image, the first relative pose data, the second image, and the second relative pose data.

2. The method of claim 1 , further comprising:

generating first relative pose information based on the first relative pose data of the object, and storing the first image and the first relative pose information as training data or transmitting the first image and the first relative pose information to a server communicating with the computing device; and

generating second relative pose information based on the second relative pose data of the object, and storing the second image and the second relative pose information as training data or transmitting the second image and the second relative pose information to the server communicating with the computing device.

3. The method of claim 2 , further comprising receiving size information of a virtual block corresponding to the object, wherein the generating of the first relative pose information comprises:

generating a virtual block coordinate on the object coordinate system, based on the size information of the virtual block;

generating a virtual block coordinate on the first camera coordinate system, based on the virtual block coordinate on the object coordinate system and the first relative pose data of the object; and

generating, as the first relative pose information, a virtual block coordinate on a two-dimensional camera coordinate system at the first time point, based on the virtual block coordinate on the first camera coordinate system.

4. The method of claim 1 , wherein the object is arranged in a preset direction at a preset position with respect to the marker.

5. The method of claim 1 , wherein the generating of the second view data of the camera comprises:

estimating a direction and magnitude of a translational motion and a direction and magnitude of a rotational motion in response to a motion of the camera between the first time point and the second time point; and

generating the second view data of the camera based on the first view data of the camera, the direction and magnitude of the translational motion, and the direction and magnitude of the rotational motion.

6. The method of claim 5 , wherein the direction and magnitude of the translational motion and the direction and magnitude of the rotational motion are estimated based on a change between feature points of the first image and feature points of the second image.

7. The method of claim 5 , wherein the direction and magnitude of the translational motion and the direction and magnitude of the rotational motion are estimated based on a change between feature points of the first image and feature points in the second image and a sensor value of an inertial sensor in the camera.

8. The method of claim 1 , wherein:

the second view data of the camera is a matrix for converting a coordinate value on the world coordinate system onto a second camera coordinate system defined by a pose of the camera at the second time point.

9. The method of claim 8 , wherein:

the second relative pose data of the object is a matrix for converting a coordinate value on the object coordinate system onto the second camera coordinate system.

10. The method of claim 9 , wherein the estimating of the pose data of the object comprises:

calculating first pose data representing a pose of the camera at the first time point on the world coordinate system, based on the first view data of the camera; and

calculating the pose data of the object by multiplying the first pose data by the first relative pose data of the object.

11. The method of claim 9 , wherein the second relative pose data of the object is calculated by multiplying the second view data of the camera by the pose data of the object.

12. The method of claim 1 , wherein the first view data of the camera, the first relative pose data of the object, and the pose data of the object are each in a form of a 4×4 pose matrix including a 3×3 rotation matrix and a 3×1 translation matrix.