Concept for an entry-exit matching system
Examples relate to a concept for an entry-exit matching system, and in particular to an evaluation device, a method and a computer program for person re-identification for entry-exit matching in a transportation system. The evaluation device comprises processing circuitry configured to obtain a plurality of re-identification codes. Each re-identification code represents a person being recorded by at least one camera when entering or exiting at least a section of the transportation system. The processing circuitry is configured to match the plurality of re-identification codes using a global matching scheme to obtain a plurality of matched pairs of re-identification codes, such that each matched pair of re-identification codes comprises a re-identification code of a person entering and a re-identification code of a person exiting. The global matching scheme is based on reducing an overall distance between the re-identification codes of the matched pairs of re-identification codes over the plurality of matched pairs of re-identification codes. The processing circuitry is configured to determine points of entry and exit for the plurality of matched pairs of re-identification codes.
1 . An evaluation device for person re-identification in a transportation system, the evaluation device comprising processing circuitry configured to:
obtain a plurality of re-identification codes,
wherein each re-identification code represents a person being recorded by at least one camera when entering or exiting respective points of entry and exit of at least a section of the transportation system,
match the plurality of re-identification codes using a global matching scheme to obtain a plurality of matched pairs of re-identification codes, such that each matched pair of re-identification codes comprises a re-identification code of a person entering and a re-identification code of a person exiting,
wherein the global matching scheme comprises solving a combinatorial optimization problem across the plurality of re-identification codes to minimize a total cost function defined over all pairs, wherein the total cost function is further adapted based on prior statistical knowledge on a plurality of journeys between the points of entry and exit; and
determine matched points of entry and exit for the plurality of matched pairs of re-identification codes.
2 . The evaluation device according to claim 1 , wherein the global matching scheme is based on a combinatorial optimization algorithm.
3 . The evaluation device according to claim 1 , wherein the global matching scheme is based on a graph-based algorithm.
4 . The evaluation device according to claim 1 , wherein each pair of two re-identification codes is associated with a cost value that is based on the distance between the re-identification codes of the pair, wherein the global matching scheme is based on reducing the overall sum of the cost values of the plurality of matched pairs of re-identification codes.
5 . The evaluation device according to claim 1 , wherein each re-identification code is further based on one or more of facial features of the respective person, a gait of the person, an estimated age of the person, an estimated gender of the person, an estimated height of the person, an estimated length of body parts, and a clothing of the person.
6 . The evaluation device according to claim 1 , wherein each re-identification code is associated with a timestamp and/or location information, wherein the processing circuitry is configured to determine the time and/or location of entry and exit for the plurality of matched pairs of re-identification codes.
7 . The evaluation device according to claim 1 , wherein the processing circuitry is configured to update the matching of the plurality of re-identification codes based on the global matching scheme when a re-identification code is added to the plurality of re-identification codes.
8 . The evaluation device according to claim 1 , wherein the processing circuitry is configured to obtain image data of the at least one camera, and to generate the plurality of re-identification codes based on the image data.
9 . The evaluation device according to claim 8 , wherein the processing circuitry is configured to track persons over a plurality of frames of image data, to determine that a person is entering or exiting the transportation system based on the tracking of the person over the plurality of frames of image data, and to generate a re-identification code of the person upon determination of the person entering or exiting the transportation system.
10 . The evaluation device according to claim 1 , wherein the processing circuitry is configured to track persons over a plurality of frames of image data, and to generate the re-identification code of the person based on one of the frames of image data based on a suitability of the respective frame for the generation of a re-identification code.
11 . The evaluation device according to claim 10 , wherein the processing circuitry is configured to determine the suitability of the frames using a clustering algorithm.
12 . The evaluation device according to claim 10 , wherein the processing circuitry is configured to determine the suitability of the frames based on an angle of a face of the person relative to the at least one camera.
13 . The evaluation device according to claim 1 , wherein the transportation system is a single vehicle, wherein each re-identification code represents a person being recorded by a camera when entering or exiting the vehicle.
14 . A method for person re-identification in a transportation system, the method comprising:
obtaining a plurality of re-identification codes,
wherein each re-identification code represents a person being recorded by at least one camera when entering or exiting respective points of entry and exit of at least a section of the transportation system;
matching the plurality of re-identification codes using a global matching scheme to obtain a plurality of matched pairs of re-identification codes, such that each matched pair of re-identification code comprises a re-identification code of a person entering and a re-identification code of a person exiting,
wherein the global matching scheme comprises solving a combinatorial optimization problem across the plurality of re-identification codes to minimize a total cost function defined over all pairs, wherein the total cost function is further adapted based on prior statistical knowledge on a plurality of journeys between the points of entry and exit; and
determining matched points of entry and exit for the plurality of matched pairs of re-identification codes.
15 . A non-transitory, computer-readable medium comprising a program code that, when the program code is executed on a processor, a computer, or a programmable hardware component, causes the processor, computer, or programmable hardware component to perform the method of claim 14 .
16 . An evaluation device for person re-identification in a transportation system, the evaluation device comprising processing circuitry configured to:
obtain a plurality of transformed re-identification codes,
wherein each transformed re-identification code is based on a similarity-preserving transformation of a re-identification code using a transformation parameter that is dependent on at least one of a time and a location, and
wherein each re-identification code represents a person being recorded by at least one camera when entering or exiting respective points of entry and exit of at least a section of the transportation system,
match the plurality of transformed re-identification codes that are based on the transformation parameter using a global matching scheme to obtain a plurality of matched pairs of transformed re-identification codes, such that each matched pair of transformed re-identification codes comprises a transformed re-identification code of a person entering and a transformed re-identification code of a person exiting,
wherein the global matching scheme is based on reducing an overall distance between the transformed re-identification codes of the matched pairs of transformed re-identification codes over the plurality of matched pairs of transformed re-identification codes and further based on prior statistical knowledge on the points of entry and exit; and
determine matched points of entry and exit for the plurality of matched pairs of transformed re-identification codes.