Real-Time Pedestrian Tracking and Counting with TLD

Algorithm 1

The flow for multiobject tracking.

Input: Detected targets in current frame j; Tracking list ;

   Current bounding box set given by TLD

Output: updated tracking list

1 Trajectory Initialization. Initialize T according . If frame number is 1 then return

after initialization.

2 Correlation computation. For each trajectory , compute the distance similarity to each

patch in the detection set , then find its most similar patch according (9).

3 Trajectory Updating. If is found, then update the trajectory based on

according to (10).

Else, update some parameters of trajectory according to .

4 Trajectory Deletion. Search throughout the updated trajectory set T. If a trajectory

satisfies following condition, then delete it from T.

5 Trajectory Insertion. For those unmatched patches in , insert them into

to form new trajectory records. Suppose the index number is m in T for a new record. Then the

parameters for are set as following.