Abstract: The present disclosure generally relates to video surveillance systems and optionally computer-implemented video management methods for video surveillance systems. A video management system may be configured to display search results of a plurality of video streams as respective thumbnails on a geo-map at respective positions of the search results within a surveillance area.

Abstract: A method of obtaining an attention matrix for use in a transformer-based model, a non-transitory computer readable storage medium, and an apparatus are provided. The method includes applying an attentional mechanism over a quadtree representation of an input image. The method may be used for image classification, for example in the field of video surveillance.

Abstract: A system and method for improving operation of a physical security event processing system by generating and deploying event-specific alarm response agent applications. The method includes: receiving a human-provided rationale for non-resolution or resolution of an alarm; generating a candidate alarm-response policy using a language model and/or a vision-language model; deriving machine representations from event data including video analytics metadata, embeddings, and/or knowledge graph context; validating reproducibility of the rationale by computing similarity or pattern matches between stored reference representations and representations derived from subsequent events; and controlling an alarm subsystem by suppressing an alarm or creating an alarm when validated reproducibility conditions are satisfied, thereby reducing false alarms and reducing computing and network load.

Abstract: A video surveillance system comprising a video management system, one or more computer-implemented digital devices and a plurality of application programs, the system being configured to run the programs on one of or across several of the said one or more computer-implemented digital devices, and configured to perform parallel processing of one or more groups of pictures from one or more video streams with the plurality of application programs in one or across several of the said one or more computer-implemented digital devices.

Abstract: A system and method for generating federated machine learning (ML) model are provided herein. The method may include: distributing, by a software agent executed by a computing device, a first machine learning model to a plurality of user devices, wherein the software agent provides the plurality of user devices with instructions for the training of the first ML model; each of the plurality of user devices, training the first ML model individually by submitting to the first ML model labelled user data sets, wherein the training of the ML model proceeds according to the instructions provided by the software agent; receiving, by the computing device, the individually trained first ML models, and identifying whether the ML models have been trained according to the software agent's instructions; and combining the individually trained first ML models to create a federated ML model which has been trained according to the software agent's instructions.

Abstract: A system and method for video selection and labelling including: receiving from the human operator a selection of first data items from the plurality of data items within the first video assembly; identifying second video assemblies comprising a sequence of frames and second data items, wherein the second data items are similar to the selection of the first data items; generating one or more candidate video assemblies comprising a sequence of frames, wherein the sequence of frames comprises a subset of the frames and data items from the second video assemblies which have been identified to be similar to the first data items, and a natural language label describing the sequence of frames; and presenting the human operator the one or more candidate video assemblies to receive feedback on the generated labels.

Abstract: The present disclosure generally relates to video surveillance systems, video management systems and optionally computer-implemented video management methods that utilize trail mapping of one of movable video cameras onto a geo-map of a surveillance area.

Abstract: A video surveillance system comprising an operator client including a system memory, a plurality of display devices for displaying video data and resources for decoding of encoded video data for display, including a plurality of GPUs, wherein each of the display devices is connected to a graphics output port of one of the GPUs. A load distribution module is configured to send a stream of encoded video data to one of the decoding resources by determining a first display device the stream of video data is to be displayed after decoding, and sending the stream of video data to a first GPU for decoding and display on the selected display device, wherein the first GPU is connected to the first display device, unless the first GPU is overloaded. If the first GPU is overloaded, the load distribution module sends the stream of video data to a second GPU for decoding.

Abstract: The present disclosure aims to allow users such as Health Care Personnel, nurses or more generally speaking video surveillance operators to blur video surveillance data and/or attach content such as comments or notes to video surveillance data at a client-level.

Abstract: The present disclosure generally relates to video surveillance systems and optionally computer-implemented video management methods for video surveillance systems. A video management system may be configured to dynamically calculate the Field-of-View (FOV) of a video camera using Geographic Information System (GIS) data.

Abstract: A video surveillance system including a movable video camera, a plurality of stationary video cameras, wherein the movable video camera is geographically dynamic while recording or streaming video, and a video management system comprising a processing unit, and a user screen or display comprising a user interface window. The processing unit is configured to determine a geographical position and a time instant of the movable video camera, and respective geographical positions of the plurality of stationary video cameras, select a subset of stationary video cameras arranged nearby the movable video camera, present a video stream of the movable video camera in a primary tile, present the respective video streams of the subset of stationary video cameras in respective secondary tiles of the user interface window.

Abstract: A computer-implemented method of generating a thumbnail of a video stream or file of a surveillance area in a video management system, including setting, in the video management system, at least one sound event to be detected in at least one audio stream or file corresponding to the video stream or file; detecting, in the at least one audio stream or file, at least one point in time at which the at least one sound event occurs; generating the thumbnail based on at least part of at least one frame of the video stream or file, wherein the frame is selected based on the point in time; and displaying the thumbnail in the video management system.

Abstract: A video processing apparatus configured to process a stream of video surveillance data, wherein the video surveillance data includes metadata associated with video data, the metadata describing at least one object in the video data. The apparatus comprises means for applying an image assessment algorithm to generate a reliability score for the metadata, and associating the reliability score with the metadata. The image assessment algorithm generates the reliability score based on an assessment of the image quality of the video data to which the metadata relates to indicate a likelihood that the metadata accurately describes the object. An image enhancement module applies image enhancement to video data if the reliability score of metadata associated with the video data indicates a low likelihood that the metadata accurately describes the object.

Abstract: A method of training a machine learning algorithm to identify objects or activities in video surveillance data comprises generating a 3D simulation of a real environment from video surveillance data captured by at least one video surveillance camera installed in the real environment. Objects or activities are synthesized within the simulated 3D environment and the synthesized objects or activities within the simulated 3D environment are used as training data to train the machine learning algorithm to identify objects or activities, wherein the synthesized objects or activities within the simulated 3D environment used as training data are all viewed from the same viewpoint in the simulated 3D environment.

Abstract: A computer-implemented method of video surveillance comprising the steps of: querying a knowledge graph representing a plurality of video cameras as ontology entities connected by edges, in order to identify, based on a first video camera providing a first video stream that can be fed to at least one first application program to obtain first analytics data, one or more second video cameras which can each provide a respective second video stream; and identifying, based on the first application program and/or based on the first analytics data, at least one second application program that can be fed with at least one second video stream to obtain second analytics data.

Abstract: A computer implemented method of training an operator of a video management system, the method comprising inputting, into the video management system, a plurality of virtual video streams generated by virtual video cameras within a dynamic 3D virtual environment; displaying on a display, from the virtual management system, at least some of the said virtual video streams, wherein at least one output video stream shows an event of interest; receiving, in the computer, at least one command from the operator indicative of a detection of the event of interest in the dynamic 3D virtual environment by the operator.

Abstract: A video surveillance system comprising a video management system and one or more digital devices, each of the one or more digital devices being configured to emulate at least one physical video camera, and to send video streams and/or video metadata via a computer network to a video-receiving system, wherein the video management system comprises: an input interface for receiving one or more video streams from one or more video cameras and/or other video sources, a processing unit configured to receive one or more input video streams from the input interface, each input video stream corresponding to one of the received video streams, and to store the input video streams in a video repository, and an output interface configured to send, via a computer network, one or more of the input video streams and/or one or more of the stored video streams to one or more of the digital devices.

Abstract: A computer implemented method of operating a video management system, the method comprising inputting, into the video management system, a plurality of virtual video streams generated by virtual video cameras within a dynamic 3D virtual environment; receiving, in the computer, at least one command from a user indicative of an instruction to alter a functioning of the dynamic 3D virtual environment and/or a functioning of the video management system; and displaying on a display, from the video management system, at least one virtual video stream and/or at least one alert, alarm and/or message implementing at least one alteration caused by the received command.

Abstract: A method for setting parameters values of a video source device comprising obtaining a value of an image characteristic for a current image generated by the video source device set with initial parameters values; determining a satisfaction level for the image characteristic of the current image based on the obtained value, the satisfaction level representing a probability to fulfil a task; obtaining a set of candidates, a candidate being defined as a set of parameters values; for each candidate, predicting an evolution of the satisfaction level for the image characteristic relatively to the satisfaction level determined while the video source device is set with the initial parameters values; selecting a candidate based on its predicted evolution of the satisfaction level; setting the parameters values of the video source device using the set of parameters values of the selected candidate.

Abstract: A method for setting a parameter value of a first video source device comprises: determining a first set of image characteristic values from an image captured by the first video source device configured with a first set of parameter values; obtaining a second set of image characteristic values generated from an image captured by a second video source device different from the first video source device, the second video source device being configured with a second set of parameter values; determining if the obtained second set of image characteristic values comprises values equivalent to values of the determined first set of image characteristic values; and if there are equivalent values, setting the parameter value of the first video source device using the second set of image characteristic values.