Bull Sas

Commissariat &#xc0; L'&#xc9;nergie Atomique Et Aux &#xc9;nergies Alternatives

Benoit Pelletier

Cécile Boukamel-Donnou

Rémi Druihle

Marc Platini

Malo Jennequin

Loïc Albertin

Nicolas Arnaise

Mathieu Ospici

Loic Pauletto

Jullian Bellino

Disclosed is a method for training a neural network, for the recognition of a sequence of characters in an image and without a predefined format, including: a step of creating an artificial database of a plurality of sequences of characters in images, some sequences being generated randomly, other sequences being derived from transformations of sequences generated randomly; a learning step that teaches the neural network to recognize a sequence of characters in an image, directly on at least some of the sequences of the artificial database, without a preliminary subdividing of these sequences into their component characters.

Method for training a neural network for recognition of a character sequence and associated recognition method

An electronic device for identifying and tracking at least one target, including: a receiver module configured to receive images from a network of image sensors; a determination module configured to determine visual signatures of a respective target from the received images, each visual signature being determined from a respective image including a representation of the target, each visual signature being in the form of a vector in multi-dimensional space; a first filtering module configured to filter the visual signatures; a computation module configured to compute a reference visual signature, in multi-dimensional space; a detection module configured to detect an appearance of the target, and a second filtering module.

Electronic device for identifying and tracking at least one target, associated electronic system, method and computer program

The invention relates to a computer network comprising a group of multiple computing resource infrastructures (to) associated with multiple orchestrators (to), said orchestrators being responsible for allocating the resources of this infrastructure (to) to one or more additional components of an already-hosted client application () and being grouped together in a swarm in which they are interconnected by a cooperation interface (). The allocation of resources is decided by a decision method based first on evaluations distributed to the orchestrators (to), then on a consensus protocol, between the orchestrators (to), which is based on the evaluations and which is implemented at the cooperation interface () in order to select one of the infrastructures (to) from the group to host the additional component(s) of the client application ().

Computer network of computing resource infrastructures and method for allocating said resources to client applications

The method includes for each of a plurality of successive images of a video stream from a camera, the search for at least one person present in the image and the definition, for each person found, of a zone in the image, known as person zone, surrounding this person at least partially; for each of at least one person, the grouping together into one tracklet of several person zones from successive images and surrounding this same person at least partially; for each tracklet: the identification of the person in this tracklet from person zones, the determination of a moment at which the line is crossed by the person identified from person zones and the addition of the name found and of the moment of crossing determined in at least some of the images containing the person zones.

Method for video surveillance of the crossing of a line by people, associated computer program and device

A method includes for each of a plurality of successive images in a camera video stream, searching for at least one person present in the image and defining, in the image, for each person found, a field, called person field, at least partially surrounding that person; for each of at least one person found, gathering into a track segment several person fields derived from successive images and at least partially surrounding that same person; for each track segment, identifying the person in that track segment, by a visual signature from that person, this identification including: for each person field in the track segment, determining a visual signature from the person in that track segment, called local visual signature; determining an aggregated visual signature from the local visual signatures; and identifying the person in that track segment from the aggregated visual signature.

Growing community of inventors

Saint-Étienne-de-Crossey, France

Benoit Pelletier