Download presentation
Presentation is loading. Please wait.
1
Tracking Moving Objects in Anonymized Trajectories Nikolay Vyahhi 1, Spiridon Bakiras 2, Panos Kalnis 3, and Gabriel Ghinita 3 1 St. Petersburg State University 2 John Jay College, City Univ. of New York 3 National University of Singapore
2
2 Motivation Collection of Trajectory Data Example: Traffic monitoring system GPS or Sensors deployed across a city Queries: Predict traffic conditions Data expected to be anonymous Remove ID Reconstruction of original trajectories E.g., Police tracking a suspect
3
3 Problem Statement Given a large database with anonymized spatio-temporal measurements, reconstruct the original object trajectories Requirements Efficiency (large databases) Accuracy (useful results)
4
4 Problem Statement Input: A series of M snapshots S i, each containing exactly N measurements from timestamp t i Output: A set of N trajectories Each measurement can be associated with a single trajectory M = N = 3
5
5 Related work: Multiple Target Tracking This problem is closely related to multiple target tracking (MTT) algorithms Studied in the field of radar technology Three major categories Nearest neighbor (NN) Joint probabilistic data association (JPDA) Multiple hypothesis tracking (MHT)
6
6 Related work: NN and JPDA They work in a single scan of the dataset Greedy approach: in each timestamp, every sample is associated with a single track Objective: minimize the error across all associations in the current timestamp Performance: Efficient – can work in polynomial time Greedy approach results in many false associations
7
7 Related work: MHT Multiple hypotheses are maintained Joint probabilities are calculated recursively when new measurements are received Each association is based on both previous and subsequent data (multiple scans) Unfeasible hypotheses are eventually eliminated Performance: Very accurate Computational and space complexity is exponential to the number of measurements
8
8 Comparison Very accurate Very slow Large errors Fast Very accurate Much faster than MHT
9
9 Our Approach MCMF: Min-cost Max-flow Transform the tracking problem into a min-cost max-flow problem Min-cost max-flow (graph algorithm) Input: a weighted graph G with two special nodes (source s and destination t) Objective: find the maximum flow that can be sent from s to t that results in the minimum cost Well-known algorithms exist that work in polynomial time
10
10 Transformation All edges have capacity 1 Node id (t i, p i, p j ): the object moves from location p i in timestamp t i to location p j in timestamp t i+1
11
11 Calculating the Cost Values Assume two successive measurements (p i and p j ) belong to the same track Use these values to predict the next location Calculate the error (i.e., cost) for every possible location p k
12
12 Limitation of this Approach Problem: A single measurement can be associated with multiple tracks!
13
13 Solution: Create a Block for each Measurement Corresponds to all partial tracks p m-1,i p m,k p m+1,j A block containing a flow is marked as active The only possible route inside an active block, is through the reverse path of the existing flow Block for k th measurement of m th timestamp (p m,k )
14
14 Block Functionality Block for p 3,1 Block for p 2,1 Original track: p 2,1 p 3,1 p 4,1 New track: p 2,2 p 3,1 p 4,1 Original track: p 1,1 p 2,1 p 3,1 New track: p 1,1 p 2,1 p 3,2
15
15 Improving the Running Time Flow network is too large Inefficient, since solution requires multiple shortest path calculations Assume any object can travel at most R max distance between two consecutive timestamps. R max depends on The maximum speed of the objects The time interval between two timestamps This reduces significantly the number of vertices and edges inside each block
16
16 The Tracking Algorithm Successive Shortest Path Algorithm At each iteration, send a single flow unit across the shortest path from s to t Total of N iterations in our case Most efficient implementation: Dijkstra with Fibonacci heap for priority queue Graph contains negative weights, but can utilize vertex potentials to avoid this (provided that there are no negative weight cycles) Bellman-Ford also works very well
17
17 Dealing with Negative Weight Cycles Negative weight cycles do appear in MCMF calculations In this case, follow a greedy approach: Output all the tracks that are discovered so far they might not be optimal Remove all vertices and edges associated with these tracks from the flow network Start a new min-cost max-flow calculation on the reduced graph
18
18 Complexity Computational: N iterations of a shortest path algorithm O(MN 2 K(log(MNK) + K)) for Dijkstra with Fibonacci heap K is the average number of feasible associations (due to R max ) per measurement Space: O(MNK 2 ) for storing the graph
19
19 Experimental Evaluation Data generator: Road map of San Francisco city For each object, randomly select a starting point and a destination point The object then follows the shortest path between the two points At each timestamp, every object i covers a distance d i [0,R max ] Number of measurements: 50,000 to 500,000
20
20 Experimental Evaluation Competitor: Global Nearest Neighbor (GNN) Employs clustering within each snapshot Considered the best single scan algorithm – runs in O(MNC 2 ) time (C is the average cluster size) Performance metrics: CPU time Success rate – percentage of partial tracks (triplets) that agree with original data
21
21 Variable N CPU time [sec] Success rate [%]
22
22 Variable R max (speed) CPU time [sec] Success rate [%]
23
23 Points to Remember Multiple-Target Tracking Large Anonymized Trajectory Databases Existing methods are either inefficient or inaccurate We proposed a polynomial time solution based on a novel transformation of the MTT problem into a min-cost max-flow problem Very accurate Need to improve the running time
24
24 Bibliography on LBS Privacy http://anonym.comp.nus.edu.sg
Similar presentations
© 2024 SlidePlayer.com Inc.
All rights reserved.