Human tracking and counting using the KINECT range sensor based on Adaboost and Kalman Filter
description
Transcript of Human tracking and counting using the KINECT range sensor based on Adaboost and Kalman Filter
Human tracking and counting using the KINECT range sensor based on Adaboost and Kalman Filter
ISVC 2013
Problem
• Human tracking
Avoid occlusion
Human Detection
• Observations:– There is an empty space in the front and back of
head– The right side of right shoulder and the left side of
left shoulder are also empty– There is a height difference between the head and
the two shoulders
How to describe the spatial information of 3D HASP
• Those criteria can be formulated as the difference of two pixel areas in the depth map – Haar-like feature
• Adaboost is introduced to construct a strong classifiers from those weak criteria
Human Detection
Human Detection by Adaboost
• Framework
Spatial feature
• Processing window– 20 redefined sub-windows
Spatial feature
• Four Haar-like features
Depth integral image
• The sum of rectangle pixel values from the top-left corner to a pixel in depth image– To speed up the computation of Haar-like features
• All pixel intensity values of D:( ) (4) (3) (2) (1)areaValue D dd dd dd dd
Adaboost algorithm
• Construct a strong classifier by a weighted linear combination of weak classifiers
1, * ( ) *
0, * ( ) *
1, * ( )H( , , , )
1,
j j
j j
H jF
H j
wherep h x p
h x potherwise
Our Classifier
• Challenge– Human can stand and face all directions with many
postures
• Solutions– Combine a horizontal strong classifier and a
vertical strong classifier
( ) ( ) | ( )C hor verwin win winF F F
Horizontal Strong Classifier
• Formulation
1, * ( ) *( )
0, * ( ) *j j
horj j
H jwin
H jF
Vertical Strong Classifier
• Formulation
1, * ( ) *( )
0, * ( ) *j j
verj j
H jwin
H jF
Training
• Took many depth maps of each object by rotating a certain degree
• 720 positive images + 288 negative images
Results
• Testing on three datasets:– Dataset 1: only one human object standing in
different directions– Dataset 2: Two human objects– Dataset 3: three or more human objects
Results (Dataset 1)
Results (Dataset 2)
Results (Dataset 3)
Choice of window sizes
Limitation
• Fails if detected humans are standing two very close to each other– Improve tracking accuracy by incorporating
Kalman Filter, since the closing time is short in real tracking application.
Conclusion
• We construct a real-time human detection based the depth image from Kinect sensor
• Head and Shoulder Profile described by some Haar-like features is incorporated into Adaboost algorithm to detect human objects.
• Detection time for each image is about 33 ms.