X-Git-Url: https://www.fleuret.org/cgi-bin/gitweb/gitweb.cgi?p=mtp.git;a=blobdiff_plain;f=README.txt;h=afc2388f86a2424ba86035f237129a094de599a0;hp=23a7ae3fb05baf63e0dbf294f6e2e0ecf048be41;hb=20f94c6bff1541da8d0639966787dc8aaafea025;hpb=332b24fec902d44c6a75a14c023ec789bd4dfba0 diff --git a/README.txt b/README.txt index 23a7ae3..afc2388 100644 --- a/README.txt +++ b/README.txt @@ -1,30 +1,134 @@ -This is a very simple implementation of the KSP applied to -multi-target tracking. It is dubbed Multi-Tracked Path. -The two main classes are MTPGraph and Tracker. + Multi-Tracked Paths (MTP) + ------------------------- -MTPGraph allows to define a directed acyclic graph (DAG), to associate -a length to each of its edge (which can be negative), and to compute -the family of paths in this graph that minimize the sum of the length -of their edges. +* INTRODUCTION -Tracker allows +This is a very simple implementation of a variant of the k-shortest +paths algorithm (KSP) applied to multi-target tracking, as described +in - (1) to define a spatial topology composed of + J. Berclaz, E. Turetken, F. Fleuret, and P. Fua. Multiple Object + Tracking using K-Shortest Paths Optimization. IEEE Transactions on + Pattern Analysis and Machine Intelligence (TPAMI), 33(9):1806-1819, + 2011. + +This implementation is not the reference implementation used for the +experiments presented in this article. + +* INSTALLATION + +This software should compile with any C++ compiler. Under a unix-like +environment, just execute + + make + ./mtp_example + +It will create a synthetic dummy example, save its description in +tracker.dat, and print the optimal detected trajectories. + +If you now execute + + ./mtp tracker.dat + +It will load the file tracker.dat saved by the previous command, run +the detection, save the detected trajectories in result.trj, and the +underlying graph with occupied edges in graph.dot. + +If you do have the graphviz set of tools installed, you can produce a +pdf from the latter with the dot command: + + dot < graph.dot -T pdf -o graph.pdf + +* IMPLEMENTATION + +The two main classes are MTPGraph and MTPTracker. + +The MTPGraph class contains a directed acyclic graph (DAG), with a +length for each edge -- which can be negative -- and has methods to +compute the family of paths in this graph that globally minimizes the +sum of edge lengths. + +If there are no path of negative length, this optimal family will be +empty, since the minimum total length you can achieve is zero. Note +that the procedure is similar to that of KSP, in the sense that the +family it computes eventually is globally optimal, even if the +computation is iterative. + +The MTPTracker is defined by + + (1) a spatial topology composed of - a number of locations - - the allowed motions between them (i.e. a Boolean flag for each - pair of locations) - - the entrances (a Boolean flag for each location) - - the exits (a Boolean flag for each location) - (2) to define a number of time steps + - the allowed motions between them (a Boolean flag for each pair + of locations from/to) + + - the entrances (a Boolean flag for each location and time step) + + - the exits (a Boolean flag for each location and time step) + + (2) a number of time steps + + (3) a detection score for every location and time, which stands for + log(P(Y = 1 | X)/P(Y = 0 | X)) where Y is the said location + occupancy and X the available observations. + +From this setting, MTPTracker has methods to compute the best set of +disjoint trajectories consistent with the topology, which maximizes +the overall detection score (i.e. the sum of the detection scores of +the nodes visited by the trajectories). If no trajectory of total +positive detection score exists, this optimal set of trajectories will +be empty. + +The MTPTracker is a wrapper around the MTPGraph class. + +From the defined spatial topology and number of time steps, it builds +a graph with one source, one sink, and two nodes per location and +time. This structure ensures that the trajectories computed by the +MTPTracker will be node-disjoint, since the trajectories computed by +the MTPGraph are edge-disjoint. + +The edges from the source or to the sink, or between these pairs of +nodes, are of length zero, and the edges between the two nodes of such +a pair have negative lengths, equal to the opposite of the +corresponding detection scores. + +The file mtp_example.cc gives a very simple usage example of the +MTPTracker class by setting the tracker parameters dynamically, and +running the tracking. + +The tracker data file for MTPTracker::read has the following format, +where L is the number of locations and T is the number of time steps: + + ---------------------------- snip snip ------------------------------- + int:L int:T + + bool:allowed_motion_from_1_to_1 ... bool:allowed_motion_from_1_to_L + ... + bool:allowed_motion_from_L_to_1 ... bool:allowed_motion_from_L_to_L + + bool:entrance_1_1 ... bool:entrance_1_L + ... + bool:entrance_T_1 ... bool:entrance_T_L + + bool:exit_1_1 ... bool:exit_1_L + ... + bool:exit_T_1 ... bool:exit_T_L + + float:detection_score_1_1 ... float:detection_score_1_L + ... + float:detection_score_T_1 ... float:detection_score_T_L + ---------------------------- snip snip ------------------------------- - (3) to set for every location and time a detection score +The method MTPTracker::write_trajectories writes first the number of +trajectories, followed by one line per trajectory with the following +structure -From this input, it computes the best set of disjoint trajectories -consistent with the topology, which maximizes the overall detection -score (i.e. the sum of the detection scores of the nodes visited by -the trajectories) + ---------------------------- snip snip ------------------------------- + int:traj_number int:entrance_time int:duration float:score int:location_1 ... int:location_duration + ---------------------------- snip snip ------------------------------- -The file mtp.cc gives a very simple example. +-- +François Fleuret +September 2012