DETECTING MULTIPLE MOVING OBJECTS IN CROWDED ENVIRONMENTS WITH COHERENT MOTION REGIONS
First Claim
1. An apparatus for identifying at least one discrete moving object within a series of video images, said apparatus comprising:
- an image recording device that stores a time series of video images; and
an image analysis device that is configured to execute a program of machine-executable instructions to identify at least one moving object in said time series of video images, wherein said program includes the steps of;
identifying a plurality of feature point tracks in said time series of video images, wherein for each video image of said time series of video images, all feature points belonging to a same feature point track are located within a polygon of a predefined size; and
calculating a trajectory similarity factor for each pair of feature point tracks said plurality of feature point tracks, wherein each trajectory similarity factor is a measure of a maximum distance between a pair of feature point tracks.
3 Assignments
0 Petitions
Accused Products
Abstract
Coherent motion regions extend in time as well as space, enforcing consistency in detected objects over long time periods and making the algorithm robust to noisy or short point tracks. As a result of enforcing the constraint that selected coherent motion regions contain disjoint sets of tracks defined in a three-dimensional space including a time dimension. An algorithm operates directly on raw, unconditioned low-level feature point tracks, and minimizes a global measure of the coherent motion regions. At least one discrete moving object is identified in a time series of video images based on the trajectory similarity factors, which is a measure of a maximum distance between a pair of feature point tracks.
-
Citations
52 Claims
-
1. An apparatus for identifying at least one discrete moving object within a series of video images, said apparatus comprising:
-
an image recording device that stores a time series of video images; and an image analysis device that is configured to execute a program of machine-executable instructions to identify at least one moving object in said time series of video images, wherein said program includes the steps of; identifying a plurality of feature point tracks in said time series of video images, wherein for each video image of said time series of video images, all feature points belonging to a same feature point track are located within a polygon of a predefined size; and calculating a trajectory similarity factor for each pair of feature point tracks said plurality of feature point tracks, wherein each trajectory similarity factor is a measure of a maximum distance between a pair of feature point tracks. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20)
-
-
21. A method for identifying at least one discrete moving object within a series of video images, said method comprising:
-
storing a time series of video images in an image recording device; identifying a plurality of feature point tracks in said time series of video images, wherein for each video image of said time series of video images, all feature points belonging to a same feature point track are located within a polygon of a predefined size; calculating a trajectory similarity factor for each pair of feature point tracks in said plurality of feature point tracks, wherein each trajectory similarity factor is a measure of a maximum distance between a pair of feature point tracks; and identifying at least one discrete moving object in said time series of video images based on said trajectory similarity factors. - View Dependent Claims (22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40)
-
-
41. A machine-readable data storage device embodying a program of machine-executable instructions to identify at least one discrete moving object, wherein said program includes the steps of:
-
identifying a plurality of feature point tracks in said time series of video images, wherein for each video image of said time series of video images, all feature points belonging to a same feature point track are located within a polygon of a predefined size; calculating a trajectory similarity factor for each pair of feature point tracks said plurality of feature point tracks, wherein each trajectory similarity factor is a measure of a maximum distance between a pair of feature point tracks; and identifying at least one discrete moving object in said time series of video images based on said trajectory similarity factors. - View Dependent Claims (42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52)
-
Specification