Alerta

INFORMATION PROCESSING DEVICE, IDENTIFICATION METHOD, AND IDENTIFICATION PROGRAM

Resumen de: WO2026191163A1

An information processing device (100) comprises: an acquisition unit (120) that acquires a video including a door of a train or a platform door of a station, and audio data associated with the video; an analysis unit (130) that uses the audio data and analyzes announcement timing and announcement information indicating the content of an announcement; a vector calculation unit (140) that calculates, on the basis of the video and the announcement timing, pre-announcement movement vectors indicating movements of people before the announcement and post-announcement movement vectors indicating movements of people after the announcement; a determination unit (150) that determines, on the basis of the pre-announcement movement vectors and the post-announcement movement vectors, whether or not a flow of people is controlled by the announcement; and an output unit (160) that, when it is determined that a flow of people is controlled by the announcement, outputs the video and the announcement information as data indicating that the flow of people is under control.

TERMINAL DEVICE AND VIDEO DATA MANAGEMENT SYSTEM

Resumen de: WO2026191052A1

This terminal device that is mounted in a vehicle and capable of communicating with a server via a communication network, including a mobile communication network, comprises: a video acquisition unit that acquires, from an imaging device, video data indicating a video recording of a travel path of the vehicle; a location acquisition unit that acquires location information indicating the location of the terminal device; a time acquisition unit that acquires time information; a first storage unit that stores the video data as a plurality of video files created by splitting the video data at intervals of a predetermined period; and a communication control unit that uses a first communication device to transmit management information including list information indicating a list of the plurality of video files, the location information, the time information, and video provider identification information identifying a video provider, who is a provider of the video data, to the server at predetermined time intervals.

METHOD OF PROCESSING ROAD TRAFFIC DATA USING A GENERATIVE PRE-TRAINED TRANSFORMER (GPT)

Resumen de: WO2026188583A1

A method of processing road traffic data using a generative pre-trained transformer GPT (20) is provided. The method comprises receiving a set of input tensors (24) as an input (22) to the GPT (20), each input tensor (24) comprising multi-modality traffic data for one or more participants of a traffic scene (10), the GPT (20) having been pre-trained on multi-modality traffic data. The method includes the GPT (20) computing traffic features (26) of the traffic scene (10) based on observed traffic states; and the GPT (20) predicting one or more future traffic states (28) from the computed traffic features (26).

QUEUE DETECTION DEVICE, WHITE CANE COMPRISING SAME, QUEUE DETECTION METHOD, QUEUE DETECTION PROGRAM, PASSING SPACE DETECTION DEVICE, PASSING SPACE DETECTION METHOD, AND PASSING SPACE DETECTION PROGRAM

Resumen de: WO2026191510A1

This queue detection device (10) comprises an image acquisition unit (11), a human body detection unit (12), a number-of-people detection unit (13), a face detection unit (14), a face direction detection unit (15), and a queue detection unit (16). The image acquisition unit (11) acquires an image that is captured in a predetermined direction. A human body detection unit (12) detects the presence or absence of a human body from the image that is acquired by the image acquisition unit (11). The number-of-people detection unit (13) detects the number of human bodies detected by the human body detection unit (12). The face detection unit (14) detects the faces of the human bodies detected by the human body detection unit (12). The face direction detection unit (15) detects the directions of the faces detected by the face detection unit (14). The queue detection unit (16) detects the presence or absence of a queue and the direction thereof on the basis of the number of human bodies detected by the number-of-people detection unit (13) and the direction of the faces of the human bodies detected by the face direction detection unit (15).

CROSSING PREDICTION SYSTEM AND CROSSING PREDICTION METHOD

Resumen de: WO2026191416A1

A crossing prediction system according to one aspect comprises: a prediction unit that uses a machine-trained prediction model to predict, from a plurality of time-series first frame images obtained by capturing an imaging region including a peripheral area of a road, whether or not a passerby appearing in the plurality of first frame images will cross the road; a determination unit that uses a plurality of second frame images including a frame image captured at a later time than the plurality of first frame images to determine whether or not the passerby has actually crossed the road; a comparison unit that compares a prediction result of the prediction unit and a determination result of the determination unit to determine whether the prediction result of the prediction unit is correct; a teacher data generation unit that generates teacher data including the plurality of first frame images and label information indicating the determination result of the determination unit, according to the determination result of the comparison unit; and a training unit that retrains the prediction model using the teacher data.

DRIVING ASSISTANCE DEVICE AND DRIVING ASSISTANCE METHOD

Resumen de: WO2026191054A1

The purpose of the present disclosure is to provide a technology pertaining to vehicle driving assistance for minimizing congestion that has already occurred. A driving assistance device according to the present disclosure comprises: a congestion determination unit (14) that determines that a target vehicle (Pa) is in a specific congestion state when a plurality of vehicles in front of and behind the target vehicle (Pa) among the nearby vehicles are in a congestion state in which the speed of travel is no higher than a reference speed, and the target vehicle (Pa) is traveling near a specific location; a resolution determination unit (15) that determines that the target vehicle (Pa) is a congestion-resolvable vehicle when the target vehicle (Pa) is in the specific congestion state and the speed of an immediately preceding vehicle is higher than the speed of the target vehicle (Pa); and an assistance unit (16) that executes a congestion resolution control in which a travel control device (25) causes the target vehicle (Pa) to accelerate so that the speed of the target vehicle (Pa), which is a congestion-resolvable vehicle, approaches the speed of the immediately preceding vehicle, or performs a notification control for causing a notification device (26) to provide a notification urging acceleration to a driver of the target vehicle (Pa).

VIDEO PROCESSING SYSTEM, VIDEO PROCESSING METHOD, AND VIDEO PROCESSING PROGRAM

Nº publicación: WO2026190951A1 17/09/2026

Solicitante:

NEC CORP [JP]
\u65E5\u672C\u96FB\u6C17\u682A\u5F0F\u4F1A\u793E

Resumen de: WO2026190951A1

A video processing system according to the present disclosure comprises: an acquisition unit that acquires one or more videos; a behavior detection unit that detects one or more behaviors of a target person appearing in at least one of the videos; a risk factor detection unit that detects, from a video, a risk factor causing an accident; a determination unit that determines, on the basis of a behavior detection result and a risk factor detection result, whether the target person avoided the accident; and an output unit that outputs a video for which the determination unit has determined that the target person avoided the accident.