Systems and methods for object tracking in videos
An object may be tracked through a video using the sizes of objects depicted in the video. The size of an object to be tracked (tracking object) may be determined using the depiction of the tracking object in a video frame. In a subsequent video frame, an object that has (1) a size that is closest to the tracking size, and (2) a size that differs from the tracking size by less than a threshold amount may be identified to be the tracking object. The tracking size may be updated with the size of the object identified to be the tracking object for use in the next video frame. This enables objects to be tracked simply and quickly by using the sizes of objects. The video frames may be framed for presentation using the positions of the tracking object in the video frames.
1 . A system for object tracking in videos, the system comprising:
one or more physical processors configured by machine-readable instructions to:
obtain video information defining a spherical video, the spherical video including video frames, the video frames depicting objects;
select a tracking object depicted in a given one of the video frames;
determine a tracking size of the tracking object depicted in the given one of the video frames;
determine sizes of the objects depicted in a subsequent video frame, the subsequent video frame depicting a first object having a given size;
identify one of the objects depicted in the subsequent video frame as the tracking object based on size differences between the tracking size of the tracking object depicted in the given one of the video frames and the sizes of the objects depicted in the subsequent video frame, wherein the first object is identified as the tracking object based on a difference between the given size and the tracking size being less than a threshold size difference, further wherein the tracking size is updated with the given size in the subsequent video frame for next identification of the tracking object; and
determine framing of the spherical video for presentation based on the identification of the tracking object in the video frames.
2 . The system of claim 1 , wherein responsive to none of the objects depicted in the subsequent video frame being identified as the tracking object:
a new tracking object is selected in the subsequent video frame; or
the identification of the tracking object for the subsequent video frame is skipped.
3 . A system for object tracking in videos, the system comprising:
one or more physical processors configured by machine-readable instructions to:
obtain video information defining a video, the video including video frames, the video frames depicting objects;
select a tracking object depicted in a given one of the video frames;
determine a tracking size of the tracking object depicted in the given one of the video frames;
determine sizes of the objects depicted in a subsequent video frame;
identify one of the objects depicted in the subsequent video frame as the tracking object based on a comparison between the tracking size of the tracking object depicted in the given one of the video frames and the sizes of the objects depicted in the subsequent video frame; and
determine framing of the video for presentation based on the identification of the tracking object in the video frames.
4 . The system of claim 3 , wherein:
the subsequent video frame depicts a first object having a given size;
the first object is identified as the tracking object based on a difference between the given size and the tracking size being less than a threshold size difference; and
the first object is not identified as the tracking object based on the difference between the given size and the tracking size being greater than the threshold size difference.
5 . The system of claim 4 , wherein responsive to none of the objects depicted in the subsequent video frame being identified as the tracking object, a new tracking object is selected in the subsequent video frame.
6 . The system of claim 4 , wherein responsive to none of the objects depicted in the subsequent video frame being identified as the tracking object, the identification of the tracking object for the subsequent video frame is skipped.
7 . The system of claim 3 , wherein a first object in the subsequent video frame with a smallest size difference with the tracking size is identified as the tracking object in the subsequent video frame.
8 . The system of claim 7 , wherein the tracking size is updated with a given size of the first object identified as the tracking object in the subsequent video frame for next identification of the tracking object.
9 . The system of claim 3 , wherein one or more of the objects depicted in the subsequent video frame are filtered out from being considered as the tracking object based on classification of the one or more objects.
10 . The system of claim 3 , wherein the determination of the framing of the video for presentation based on the identification of the tracking object in the video frames includes positioning of a viewing window for the video to include the tracking object.
11 . The system of claim 3 , wherein the video includes a spherical video.
12 . A method for object tracking in videos, the method performed by a computing system including one or more processors, the method comprising:
obtaining, by the computing system, video information defining a video, the video including video frames, the video frames depicting objects;
selecting, by the computing system, a tracking object depicted in a given one of the video frames;
determining, by the computing system, a tracking size of the tracking object depicted in the given one of the video frames;
determining, by the computing system, sizes of the objects depicted in a subsequent video frame;
identifying, by the computing system, one of the objects depicted in the subsequent video frame as the tracking object based on a comparison between the tracking size of the tracking object depicted in the given one of the video frames and the sizes of the objects depicted in the subsequent video frame; and
determining, by the computing system, framing of the video for presentation based on the identification of the tracking object in the video frames.
13 . The method of claim 12 , wherein:
the subsequent video frame depicts a first object having a given size;
the first given object is identified as the tracking object based on a difference between the given size and the tracking size being less than a threshold size difference; and
the first object is not identified as the tracking object based on the difference between the given size and the tracking size being greater than the threshold size difference.
14 . The method of claim 13 , wherein responsive to none of the objects depicted in the subsequent video frame being identified as the tracking object, the identification of the tracking object for the subsequent video frame is skipped.
15 . The method of claim 12 , wherein responsive to none of the objects depicted in the subsequent video frame being identified as the tracking object, a new tracking object is selected in the subsequent video frame.
16 . The method of claim 12 , wherein a first object in the subsequent video frame with a smallest size difference with the tracking size is identified as the tracking object in the subsequent video frame.
17 . The method of claim 16 , wherein the tracking size is updated with a given size of the first object identified as the tracking object in the subsequent video frame for next identification of the tracking object.
18 . The method of claim 12 , wherein one or more of the objects depicted in the subsequent video frame are filtered out from being considered as the tracking object based on classification of the one or more objects.
19 . The method of claim 12 , wherein determining the framing of the video for presentation based on the identification of the tracking object in the video frames includes positioning of a viewing window for the video to include the tracking object.
20 . The method of claim 12 , wherein the video includes a spherical video.