Camera motion estimation method for augmented reality tracking algorithm and system therefor

1. A camera motion estimation method for an augmented reality tracking algorithm performed by a sequence production application executed by a processor of a computing device, comprising:

displaying a target object on a sequence production interface;

setting a virtual viewpoint trajectory on the displayed target object;

generating an image sequence by rendering images obtained when the target object is viewed along the set virtual viewpoint trajectory; and

reproducing the generated image sequence,

wherein the displaying the target object on the sequence production interface comprises determining the target object, which is a pre-designed three-dimensional (3D) model corresponding to a real object, according to a user input,

further comprising setting the generated virtual image sequence as a predetermined ground-truth serving as a benchmark;

receiving a real image sequence obtained by photographing and tracking a real object corresponding to the target object from an object tracking application of a terminal; and

comparing the received real image sequence with the predetermined ground-truth to calculate an accuracy parameter of the object tracking application, and

wherein the displaying the target object on the sequence production interface comprises:

importing a pre-designed 3D model corresponding to the target object and placing the imported 3D model at an origin of 3D spatial coordinates.

2. The camera motion estimation method of claim 1 , wherein the setting the virtual viewpoint trajectory on the displayed target object comprises:

placing a virtual camera according to a user input through the sequence production interface;

moving the placed virtual camera; and

setting a trajectory of movement of the virtual camera as the virtual viewpoint trajectory.

3. The camera motion estimation method of claim 1 , wherein the setting the virtual viewpoint trajectory on the displayed target object comprises:

receiving a real viewpoint trajectory generated by photographing and tracking a real object from a camera device;

generating a virtual viewpoint trajectory corresponding to the real viewpoint trajectory; and

setting the generated virtual viewpoint trajectory as the virtual viewpoint trajectory for the target object.

4. The camera motion estimation method of claim 3 , wherein the receiving the real viewpoint trajectory generated by photographing and tracking the real object from the camera device comprises:

photographing the real object by the camera device while moving the camera device;

calculating, by a real camera of the camera device, a pose value for the real object for each viewpoint at which the real object is photographed; and

determining a real viewpoint trajectory including changes of the pose value of the real camera.

5. The camera motion estimation method of claim 4 , wherein the generating the virtual viewpoint trajectory corresponding to the real viewpoint trajectory comprises converting the pose value of the real camera for the real object into a pose value of the virtual camera for the target object.

6. The camera motion estimation method of claim 1 , wherein the generating the image sequence by rendering images obtained when the target object is viewed along the set virtual viewpoint trajectory comprises:

moving the virtual camera along the virtual viewpoint trajectory; and

generating a frame image by rendering a frame image obtained when the target object is viewed from a virtual camera pose at one viewpoint of the moving virtual camera.

7. The camera motion estimation method of claim 6 , wherein the generating the image sequence by rendering images obtained when the target object is viewed along the set virtual viewpoint trajectory comprises matching a virtual camera pose value at the time of capturing each of the rendered frame images and the corresponding frame image and including the matched virtual camera pose value and the frame image in a virtual image sequence.