An image capture device may automatically capture images. An image sensor may generate visual content based on light that becomes incident thereon. A depiction of interest within the visual content may be identified, and one or more images may be generated to include one or more portions of the visual content including the depiction of interest.

Disclosed are an electronic apparatus for recognizing a multimedia signal and an operating method of the electronic apparatus, including segmenting a detection signal into a plurality of frames; segmenting each of the frames into a plurality of blocks; and representing each of the blocks as a hash word based on a time feature and a frequency feature for each of the blocks.

A method, system and computer program product for event summarization facilitated by emotions/reactions of people near an event location is disclosed. The method includes generating a query based at least in part on reaction information and at least in part on primary video metadata. Based on the query, at least one possible event summarization match for the one or more events is retrieved from a database.

An exemplary system is configured to access data captured by one or more sensors in an industrial environment, the data comprising videos of operatives in the industrial environment. The system is further configured to determine, based on the data, a task performed by the operatives and to identify a video within the videos that represents a performance of the task comprising a variance from a reference performance of the task represented in a reference video associated with the task. The system is further configured to generate, based on the video and the variance, an output for improving the performance of the task or the reference performance of the task.

Disclosed are a video automatic editing method and system based on machine learning. The video automatic editing system based on machine learning includes at least one processor, and the at least one processor includes a video acquirer configured to acquire input video, a highlight frame extractor configured to extract at least one highlight frame from the input video using a highlight extraction model pre-trained through machine learning, and a highlight video generator configured to generate highlight video from the at least one extracted highlight frame.

A video reception unit obtains a first video and a second video that are different in an imaging position from each other. An important video determination unit determines a video having a higher degree of importance on the basis of the first video and the second video. A video adjustment report unit transmits a transmission video adjustment report that is used to adjust qualities of the first video and the second video in accordance with a result of determining a degree of importance. A first video adjustment unit adjusts the first video on the basis of the transmission video adjustment report. A second video adjustment unit adjusts the second video on the basis of the transmission video adjustment report.

Embodiments described a method of video-text pre-learning to effectively learn cross-modal representations from sparse video frames and text. Specifically, an align and prompt framework provides a video and language pre-training framework that encodes the frames and text independently using a transformer-based video encoder and a text encoder. A multi-modal encoder is then employed to capture cross-modal interaction between a plurality of video frames and a plurality of texts. The pre-training includes a prompting entity modeling that enables the model to capture fine-grained region-entity alignment.

Systems and methods for video processing are described. Embodiments of the present disclosure generate a plurality of image feature vectors corresponding to a plurality of frames of a video; generate a plurality of low-level event representation vectors based on the plurality of image feature vectors, wherein a number of the low-level event representation vectors is less than a number of the image feature vectors; generate a plurality of high-level event representation vectors based on the plurality of low-level event representation vectors, wherein a number of the high-level event representation vectors is less than the number of the low-level event representation vectors; and identify a plurality of high-level events occurring in the video based on the plurality of high-level event representation vectors.

Described herein are platforms, systems, media, and methods for providing golf swing analysis and coaching by receiving a video of an individual performing a golf swing; applying an image processing algorithm to identify a plurality of body nodes of the individual in a plurality of frames of the video and generate a virtual skeleton by connecting the plurality of body nodes in the plurality of frames of the video; identifying a plurality of key frames of the video; applying at least one mathematical swing flaw evaluation to each key frame of the video to identify any swing flaws of the individual; generating a swing score for the individual; and providing access to one or more training modules based at least on any swing flaws of the individual identified and the swing score for the individual.

Systems and methods for video analysis are provided. The systems and methods may utilize machine learning to recognize steps of a medical procedure as they are being performed, and compare them with expected steps. The systems and methods may aid in supporting a medical practitioner before the procedure, during the procedure, as well as providing feedback after the procedure has been completed.