System and method of hybrid tracking for match moving

Patent metadata

Field	Value
Publication number	US-10636152-B2
Application number	US-201715811995-A
Country	US
Kind code	B2
Filing date	Nov 14, 2017
Priority date	Nov 15, 2016
Publication date	Apr 28, 2020
Grant date	Apr 28, 2020

Abstract

A system and method that maximizes tracking speed of an object in a sequence of images by selecting a technique for tracking the object independently for each frame in a video. The system includes an object feature detector that detects object features in a reference frame of the video and a feature comparator that determines a number of object features in each frame in the sequence of images that match the detected object features in the reference frame. Moreover, a tracking pattern selector selects the type of object tracking to track the object in the current frame based on the determined matched object features between the reference frame and the current frame of the video.

First claim

What is claimed:

1. A video system for maximizing tracking speed of an object in a sequence of images, the system comprising:
a camera configured to generate a plurality of images as a sequence of images of a video;
an object feature detector configured to detect a plurality of object features in a reference frame;
a feature correspondence determiner configured to determine a number of corresponding object features in at least one image in the sequence of images of the video that match the detected object features in the reference frame; and
a tracking pattern selector configured to select one of a plurality of types of object tracking by comparing a number of the determined corresponding object features to a predetermined object pattern threshold level,
wherein the tracking pattern selector selects a first type of object tracking to track the object in the at least one image of the sequence of images when the determined number of corresponding object features does not exceed the predetermined object pattern threshold level to achieve a first tracking object speed of the object,
wherein the tracking pattern selector selects a second type of object tracking to track the object in the at least one image of the sequence of images when the determined number of corresponding object features exceeds the predetermined object pattern threshold level to achieve a second tracking object speed of the object,
wherein the detected object features in the reference frame comprises a corner pattern of the object,
wherein the second type of object tracking comprises corner based tracking of the object,
wherein the first type of object tracking comprises template based tracking of the object, and
wherein the selecting of the one type of object tracking for each image in the sequence of images of the video by the tracking pattern selector excludes the selected type of object tracking for a previous image in the sequence of images of the video.

2. The video system of claim 1 , wherein tracking pattern selector is further configured to select one of the plurality of types of object tracking separately for each image of the plurality of images in the sequence of images in the video.

3. The video system of claim 2 , wherein the tracking pattern selector selects the one type object tracking for each image independently of the selected type of object tracking for the previous image in the sequence of images in the video.

4. The video system of claim 1 , wherein tracking pattern selector is further configured to compare a number of the detected plurality of object features in the reference frame to a minimum detected feature threshold and to automatically select the first type of object tracking to track the object in the sequence of images when the number of detected object features is less than the minimum detected feature threshold.

5. The video system of claim 1 , further comprising an object position determiner configured to generate position data of the tracked object based on the selected one type of object tracking.

6. The video system of claim 5 , further comprising a current tracking output configured to output the sequence of images of the video and the generated position data for the tracked object to a digital video editor to modify the tracked object in the video using the generated position data.

7. A video system for maximizing tracking speed of an object, the system comprising:
a camera configured to generate an image frame; and
an object pattern tracker having a pattern detector configured to detect a pattern feature of an object in the image frame of a video,
wherein the object pattern tracker has a detected pattern feature processor configured to determine whether the detected pattern feature of the object exceeds a predetermined object pattern threshold level,
wherein the object pattern tracker is configured to apply a first type of video tracking to track the object in the image frame when the detected pattern feature of the object does not exceed the predetermined object pattern threshold level to achieve a first tracking object speed of the object,
wherein the object pattern tracker is configured to apply a second type of video tracking to track the object in the image frame when the detected pattern feature of the object exceeds the predetermined object pattern threshold level to achieve a second tracking object speed of the object, and
wherein the object pattern tracker applies one of the first and second types of video tracking to track the object in the image frame independently of a selected type of video tracking for a previous image frame in the video.

8. The system of claim 7 , wherein the detected pattern feature of the object comprises a corner pattern of the object.

9. The system of claim 8 , wherein the first type of video tracking comprises template based tracking of the object and the second type of video tracking comprises corner based tracking of the object.

10. The system of claim 9 , wherein the corner based tracking of the object comprises corner tracking of the object in a current image frame to generate position data of the object.

11. The system of claim 10 , wherein the template based tracking of the object comprises template tracking of the object in the current image frame to generate position data of the object.

12. The system of claim 11 , further comprising a current tracking output configured to output the image frame and the generated position data of the tracked object to a digital video editor to modify the tracked object in the image frame using the generated position data.

13. A video system for maximizing tracking speed of an object in a sequence of images by selecting a tracking type independently for each image frame in a video, the system comprising:
an object feature detector configured to detect a plurality of object features in a reference frame of the video;
a feature comparator configured to determine a number of object features in a current image frame in a sequence of images of the video that match the detected object features in the reference frame; and
a tracking pattern selector configured to select one of a plurality of types of object tracking to track an object in the current image frame based on the determined matched object features between the reference frame and the current image frame of the video,
wherein the tracking pattern selector is further configured to select the one of the plurality of types of object tracking to track the object by comparing a number of the determined matched object features to a predetermined object pattern threshold level,
wherein the tracking pattern selector selects a first type of object tracking of the plurality of types of object tracking when the number of determined matched objects does not exceed the predetermined object pattern threshold level to achieve a first tracking object speed of the object, and
wherein the tracking pattern selector selects a second type of object tracking of the plurality of types of object tracking when the number of determined matched objects exceeds the predetermined object pattern threshold level to achieve a second tracking object speed of the object.

14. The video system of claim 13 , wherein the detected object features in the reference frame comprises a corner pattern of the object.

15. The video system of claim 14 , wherein first type of object tracking comprises template based tracking of the object and the second type of object tracking comprises corner based tracking of the object.

16. The video system of claim 15 , wherein the tracking pattern selector i

Assignees

• Gvbb Holdings Sarl

Inventors

• Takada Yousuke

Classifications

• G06T2207/30201

  Face · CPC title

• H04N5/262

  Studio circuits, e.g. for mixing, switching-over, change of character of image, other special effects {; Cameras specially adapted for the electronic generation of special effects} · CPC title

• G06T7/246

  using feature-based methods, e.g. the tracking of corners or segments · CPC title

• G06T7/248

  involving reference images or patches · CPC title

• G06T2207/10024

  Color image · CPC title