Efficient multi-hypothesis multi-human 3D tracking in crowded scenes
First Claim
Patent Images
1. A method to perform multi-human three dimensional (3D) tracking, comprising:
- for each single view, providing two dimensional (2D) human detection candidates from a camera to a 2D tracking module wherein a Convolutional Neural Network (CNN) generates the 2D human detection candidates;
a. independently performing 2D tracking in each 2D tracking module and reporting promising 2D tracking hypotheses to a 3D tracking module;
b. selecting trajectories from the 2D tracking modules to generate 3D tracking hypotheses; and
c. determining a difference score between the detection and the trajectory as a weighted sum of appearance, location, blob size, and orientation.
5 Assignments
0 Petitions
Accused Products
Abstract
System and methods are disclosed to perform multi-human 3D tracking with a plurality of cameras. At each view, a module receives each camera output and provides 2D human detection candidates. A plurality of 2D tracking modules are connected to the CNNs, each 2D tracking module managing 2D tracking independently. A 3D tracking module is connected to the 2D tracking modules to receive promising 2D tracking hypotheses. The 3D tracking module selects trajectories from the 2D tracking modules to generate 3D tracking hypotheses.
-
Citations
17 Claims
-
1. A method to perform multi-human three dimensional (3D) tracking, comprising:
-
for each single view, providing two dimensional (2D) human detection candidates from a camera to a 2D tracking module wherein a Convolutional Neural Network (CNN) generates the 2D human detection candidates; a. independently performing 2D tracking in each 2D tracking module and reporting promising 2D tracking hypotheses to a 3D tracking module; b. selecting trajectories from the 2D tracking modules to generate 3D tracking hypotheses; and c. determining a difference score between the detection and the trajectory as a weighted sum of appearance, location, blob size, and orientation. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13)
-
-
14. An apparatus to perform multi-human 3D tracking with a plurality of cameras, comprising:
-
a. at each view, a module coupled to each camera to provide 2D human detection candidates; b. a plurality of 2D tracking modules each coupled to the CNN and each 2D tracking module managing 2D tracking independently; 3D tracking module coupled to the 2D tracking modules to receive promising 2D tracking hypotheses, the 3D tracking module selecting trajectories from the 2D tracking modules to generate 3D tracking hypotheses wherein a Convolutional Neural Network (CNN) generates the 2D human detection candidates and the module determining a difference score between the detection and the trajectory as a weighted sum of appearance, location, blob size, and orientation. - View Dependent Claims (15, 16, 17)
-
Specification