Abstract: A management system communicates with a moving body having a localization function. The management system acquires an image captured by a moving camera mounted on a moving body and information on a moving camera position which is a position of the moving body when the image is captured. The management system extracts an image of the target area captured by the moving camera as a target area image based on the moving camera position. The management system executes area management process for managing a target area based on the target image.

Abstract: A first world is one of a real world and a virtual world simulating the real world, and a second world is another of them. A first image is captured by a first camera in the first world, and a second image is captured by a second camera in the second world. A visual positioning system executes common processing that generates a scene graph representing a positional relationship between objects included in the image and extracts a feature amount of the scene graph. The visual positioning system performs matching between a first feature amount extracted by the common processing on the first image and a second feature amount extracted by the common processing on the second image, and then associates the first camera position in the first world and the second camera position in the second world with each other based on a result of the matching.

Abstract: A model generation method for generating a video extraction model for extracting a matching interval in a video that matches contents of an input sentence is provided. In the model generation method, a base matching interval and a sub matching interval in a training video are extracted by inputting a base sentence and a sub sentence to the video extraction model. Next, a loss for each of ground truth interval, the base matching interval, and the sub matching interval is calculated by processing a learning task of reconstructing the base sentence based on a feature value of the training video corresponding to each of them. Then, a machine learning is performed such that the first loss related to the ground truth interval is smaller than the second loss related to the base matching interval, and the second loss is smaller than the loss related to the sub matching interval.

Abstract: A tracking system for moving body includes a graph. In the graph, a node representing a single camera and a node representing a common tracking ID assigned to a moving body reflected in image data acquired by a single camera are connected via an edge. In the graph, further, nodes representing respective single cameras are connected via at least one edge representing a relationship between the at least two single cameras if there is a relationship between the at least two single cameras. In the graph, furthermore, nodes representing the at least two common tracking IDs are connected via at least one edge representing that the at least two moving bodies reflected in each video data captured by the at least two single cameras are the same moving object if the nodes representing the at least two common tracking IDs are recognized to be the same moving object.

Abstract: A re-identification system temporarily performs a re-identification process for determining whether two moving objects shown in a plurality of videos are identical or not. Similarities between the two moving objects in the re-identification process are ranked in consideration of a direction of each moving object. The rank is highest when the two moving objects are identical and the two moving objects are same in the direction. The rank is lowest when the two moving objects are not identical and the two moving objects are different in the direction. A ranking rule is that the rank is higher as the similarity is higher. The re-identification system calculates a degree of consistency between the ranking result and the ranking rule. Then, the re-identification system finally determines whether the two moving objects are identical or not based on the degree of consistency in addition to the similarities.

Abstract: Processing to generate a graph and processing to search for a tracking target by referring the graph are performed. In the processing to search for the tracking target, the feature quantity of the tracking target is extracted from the image of the tracking target. Also, a moving body having the feature quantity that is most similar to the tracking target feature quantity is specified from the moving body feature quantities extracted from at least two moving body images represented by at least two nodes constituting the graph. Then a tracking target graph including a node representing a tracking identification number assigned to the identified moving body and at least one node connected to the node representing the tracking identification number via at least one edge is specified.

Abstract: A tracking system that includes a plurality of tracking devices and a server device. Each of the tracking devices: tracks an object; transmits tracking information, which includes at least features and an identifier of the object, to nearby tracking devices and the server device; determines whether an object coincided with an object specified by the transmitted tracking information is included in the currently tracked objects; replaces the identifier of the object with the identifier included in the tracking information; and transmits the coincidence information.

Abstract: A model training method trains an object identification model that is based on machine learning. The model training method includes acquiring labeled training data where a track is given as a label to a sequence of images. The track is information representing a time series of a same moving object in the sequence of images and is automatically obtained by a tracker that tracks the same moving object in the sequence of images. The model training method further includes training the object identification model based on the labeled training data.

Abstract: A training data generation method generate labeled training data used for training an object identification model that is based on machine learning. The training data generation method includes: (A) detecting a moving object in a sequence of images; (B) tracking a same moving object in the sequence of images by using a tracker, to automatically obtain a track that is information representing a time series of the same moving object in the sequence of images; and (C) generating the labeled training data by giving the track as a label to the sequence of images.