Feature Point Identification in Sparse Optical Flow Based Tracking in a Computer Vision System

Patent metadata

Field	Value
Publication number	US-2017193669-A1
Application number	US-201615266149-A
Country	US
Kind code	A1
Filing date	Sep 15, 2016
Priority date	Dec 30, 2015
Publication date	Jul 6, 2017
Grant date	—

Abstract

A method for sparse optical flow based tracking in a computer vision system is provided that includes detecting feature points in a frame captured by a monocular camera in the computer vision system to generate a plurality of detected feature points, generating a binary image indicating locations of the detected feature points with a bit value of one, wherein all other locations in the binary image have a bit value of zero, generating another binary image indicating neighborhoods of currently tracked points, wherein locations of the neighborhoods in the binary image have a bit value of zero and all other locations in the binary image have a bit value of one, and performing a binary AND of the two binary images to generate another binary image, wherein locations in the binary image having a bit value of one indicate new feature points detected in the frame.

First claim

What is claimed is:

1 . A method for sparse optical flow based tracking in a computer vision system, the method comprising:
detecting feature points in a frame captured by a monocular camera in the computer vision system to generate a plurality of detected feature points; generating a first binary image indicating locations of the detected feature points with a bit value of one, wherein all other locations in the first binary image have a bit value of zero; generating a second binary image indicating neighborhoods of currently tracked points, wherein locations of the neighborhoods in the second binary image have a bit value of zero and all other locations in the second binary image have a bit value of one; and performing a binary AND of the first binary image and the second binary image to generate a third binary image, wherein locations in the third binary image having a bit value of one indicate new feature points detected in the frame.

2 . The method of claim 1 , further comprising determining coordinates of each new feature point from the third binary image.

3 . The method of claim 1 , wherein detecting feature points and generating a first binary image are performed by a feature point detection component of the computer vision system.

4 . The method of claim 1 , wherein detecting feature points is performed by a feature detection component of the computer vision system and generating a first binary image is performed by a new feature identification component of the computer vision system.

5 . The method of claim 1 , further comprising performing point correspondence between a new frame captured by the monocular camera and the frame based on the currently tracked points and the new feature points.

6 . The method of claim 1 , wherein the generating a second binary image is performed concurrently with the detecting feature points.

7 . The method of claim 1 , wherein the neighborhoods are one of a 3×3 or a 5×5 square of pixels.

8 . A computer vision system comprising:
a monocular camera configured to capture a two dimensional (2D) frame of a scene; a feature point detection component configured to detect a plurality of feature points in a frame; and a new feature point identification component configured to identify new feature points in the detected plurality of feature points by performing a binary AND of a first binary image and a second binary image to generate a third binary image, wherein locations in the first binary image having a bit value of one indicate locations of the detected plurality of feature points and all other locations in the first binary image have a bit value of zero, and wherein locations in the second binary image having a bit value of zero indicate neighborhoods of currently tracked feature points and all other locations in the second binary image have a bit value of one, and wherein locations of the third binary image having a bit value of one indicate new feature points.

9 . The computer vision system of claim 8 , wherein the new feature identification component is further configured to determine coordinates of each new feature points from the third binary image.

10 . The computer vision system of claim 8 , wherein the new feature identification component is further configured to generate the first binary image.

11 . The computer vision system of claim 8 , wherein the feature detection component is further configured to generate the first binary image.

12 . The computer vision system of claim 8 , further comprising a sparse optical flow component configured to perform point correspondence between a new frame captured by the monocular camera and the frame based on the currently tracked feature points and the new feature points.

13 . The computer vision system of claim 8 , further comprising a direct memory access (DMA) controller configured to generate the second binary image concurrently with detection of feature points by the feature point detection module.

14 . The computer vision system of claim 8 , wherein the neighborhoods are one of a 3×3 or a 5×5 square of pixels.

15 . A computer readable medium storing software instructions that, when executed by one or more processors comprised in a computer vision system, cause the computer vision system to execute a method for sparse optical flow based tracking, the software instructions comprising instructions to cause:
detection of feature points in a frame captured by a monocular camera in the computer vision system to generate a plurality of detected feature points; generation of a first binary image indicating locations of the detected feature points with a bit value of one, wherein all other locations in the first binary image have a bit value of zero; generation of a second binary image indicating neighborhoods of currently tracked points, wherein locations of the neighborhoods in the second binary image have a bit value of zero and all other locations in the second binary image have a bit value of one; and performance of a binary AND of the first binary image and the second binary image to generate a third binary image, wherein locations in the third binary image having a bit value of one indicate new feature points detected in the frame.

16 . The computer readable medium of claim 15 , wherein the instruction further comprise instruction to determination of coordinates of each new feature point from the third binary image.

17 . The computer readable medium of claim 15 , wherein detection of feature points and generation of a first binary image are performed by a feature point detection component of the computer vision system.

18 . The computer readable medium of claim 15 , wherein detection of feature points is performed by a feature detection component of the computer vision system and generation of a first binary image is performed by a new feature identification component of the computer vision system.

19 . The computer readable medium of claim 15 , wherein the instructions further comprise instructions to cause performance of point correspondence between a new frame captured by the monocular camera and the frame based on the currently tracked points and the new feature points.

20 . The computer readable medium of claim 15 , wherein the generation of a second binary image is performed concurrently with the detection of feature points.

Assignees

• Texas Instruments Inc

Inventors

• Poddar Deepak Kumar
• Jain Anshu
• Kumar Desappan
• Swami Pramod Kumar

Classifications

• G06T7/246 Primary

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

• G06T2207/20016

  Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform · CPC title

• G06T2207/30241

  Trajectory · CPC title

• G06T2200/28

  involving image processing hardware · CPC title

• G06T7/0042

  Physics · mapped topic