A detection apparatus includes one or more processors. The processors set at least one time-period candidate. The processors input, to a first model that inputs a feature acquired from a plurality of time-series images and the time-period candidate and outputs at least one first likelihood indicating a likelihood of occurrence of at least one action previously determined as a detection target and correction information for acquisition of at least one correction time period resulting from correction of the at least one time-period candidate, the feature and the time-period candidate, and acquire the first likelihood and the correction information output from the first model. The processors detect, based on the at least one correction time period acquired based on the correction information and the first likelihood, the action included in the time-series images and a start time and a finish time of a time period of occurrence of the action.

Systems and methods for improved operations of ski lifts increase skier safety at on-boarding and off-boarding locations by providing an always-on, always-alert system that “watches” these locations, identifies developing problem situations, and initiates mitigation actions. One or more video cameras feed live video to a video processing module. The video processing module feeds resulting sequences of images to an artificial intelligence (AI) engine. The AI engine makes an inference regarding existence of a potential problem situation based on the sequence of images. This inference is fed to an inference processing module, which determines if the inference processing module should send an alert or interact with the lift motor controller to slow or stop the lift.

A scalable tracking system processes video of a space to track the positions of objects within a space. The tracking system determines local coordinates for the objects within frames of the video and then assigns these coordinates to time windows based on when the frames were received. The tracking system then combines or clusters certain local coordinates that have been assigned to the same time window to determine a combined coordinate for an object during that time window.

A scalable tracking system processes video of a space to track the positions of people within a space. The tracking system determines local coordinates for the people within frames of the video and then assigns these coordinates to time windows based on when the frames were received. The tracking system then combines or clusters certain local coordinates that have been assigned to the same time window to determine a combined coordinate for a person during that time window.

There is provided an image processing system and method for identifying a user. The system comprises a processor configured to identify a first user in an image, determine a plurality of characteristic vectors associated with the first user, compare the characteristic vectors associated with the first user with a plurality of predetermined characteristic vectors associated with a plurality of users including the first user, and identify the first user based on the comparison.

The present application provides an action recognition method and apparatus and an electronic device. The method includes: if a target object is detected from a video frame, acquiring a plurality of images containing the target object, and optical-flow images of the plurality of images; extracting an object trajectory feature of the target object from the plurality of images, and extracting an optical-flow trajectory feature of the target object from the optical-flow images of the plurality of images; and according to the object trajectory feature and the optical-flow trajectory feature, recognizing a type of an action of the target object. Because it combines the time-feature information and the spatial-feature information of the target object, effectively increases the accuracy of the detection and recognition on the action type, and may take into consideration the detection efficiency at the same time, thereby improving the overall detection performance.

A sound effect adjustment method is provided. In the method, a video frame and an audio signal of a corresponding time unit of a target video are obtained. A sound source orientation and a sound source distance of a sound source object in the video frame are determined. Scene information corresponding to the video frame is determined. The audio signal is filtered based on the sound source orientation and the sound source distance. An echo coefficient is determined according to the scene information. Further, an adjusted audio signal with an adjusted sound effect is generated based on the filtered audio signal and the echo coefficient.

A monitoring device is configured to include an image data acquiring unit, a detecting unit, and a notifying unit. The image data acquiring unit acquires image data obtained by imaging a work area, and information relating to the position in which the image data were captured. The detecting unit detects, from the image data, at least either an unsafe behavior of a worker in the work area, or an unsafe environment around the worker, as an unsafe condition. The notifying unit notifies a terminal of the unsafe condition detected by the detecting unit, and the information relating to the position in which the unsafe condition is detected.

This invention provides a system and method that uses a hybrid model for transportation-based (e.g. maritime) visual event detection of events. In operation, video data is reduced by detecting change and exclusively transmitting images to the deep learning model when changes are detected, or alternatively, based upon a timer that samples at selected intervals. Relatively straightforward deep learning models are used, which operate on sparse individual frames, instead of employing complex deep learning models that operate on multiple frames/videos. This approach reduces the need for specialized models. Independent, rule-based classifiers are used, based on the output of the deep learning model into visual events that, in turn, allows highly specialized events to be constructed. For example, multiple detections can be combined into higher-level single events, and thus, the existence maintenance procedures, cargo activities, and/or inspection rounds can be derived from combining multiple events or multiple detections.

Embodiments provide systems and techniques for automated haptics generation and distribution. An example technique includes receiving media content from a computing device. The media content includes at least one of audio content or video content. One or more features of the media content is determined. A set of haptic data is generated for the media content, based on evaluating the one or more features of the media content with at least one machine learning model. Another example technique includes obtaining a set of haptic data associated with media content. The set of haptic data, metadata, and the media content is transmitted to a computing device.