The video processing device includes a video acquisition means, a person identification means, an importance calculation means, and an importance integration means. The video acquisition means acquires a material video. The person identification means identifies a person from the material video. The importance calculation means calculates an importance of the material video. The importance integration means integrates the importance for each person and outputs a person importance indicating an importance for each person.

Systems, methods, and non-transitory computer-readable media can obtain a test content item having a plurality of video frames. At least one video fingerprint is determined based on a set of video frames corresponding to the test content item. At least one reference content item is determined using at least a portion of the video fingerprint. At least one portion of the test content item that matches at least one portion of the reference content item is determined based at least in part on the video fingerprint of the test content item and one or more video fingerprints of the reference content item.

The present disclosure relates to systems and methods for classifying videos based on video content. For a given video file including a plurality of frames, a subset of frames is extracted for processing. Frames that are too dark, blurry, or otherwise poor classification candidates are discarded from the subset. Generally, material classification scores that describe type of material content likely included in each frame are calculated for the remaining frames in the subset. The material classification scores are used to generate material arrangement vectors that represent the spatial arrangement of material content in each frame. The material arrangement vectors are subsequently classified to generate a scene classification score vector for each frame. The scene classification results are averaged (or otherwise processed) across all frames in the subset to associate the video file with one or more predefined scene categories related to overall types of scene content of the video file.

A highlight moment within a video, music to accompany a looping presentation of the video, and a looping effect for the video may be determined. A segment of the video to be used for the looping presentation of the video may be selected based on highlight moment, the music, and the looping effect. The looping presentation of the video may be generated to have the segment edited based on a style of the looping effect and to include accompaniment of the music.

In a video processing device, a video acquisition means acquires a material video. An importance calculation means calculates importance in the material video using a plurality of models. An importance integration means integrates the importance calculated using the plurality of models. A generation means extracts important scenes in the material video based on the integrated importance and generates a digest video including the extracted important scenes.

In embodiments of event image curation, a computing device includes memory that stores a collection of digital images associated with a type of event, such as a digital photo album of digital photos associated with the event, or a video of image frames and the video is associated with the event. A curation application implements a convolutional neural network, which receives the digital images and a designation of the type of event. The convolutional neural network can then determine an importance rating of each digital image within the collection of the digital images based on the type of the event. The importance rating of a digital image is representative of an importance of the digital image to a person in context of the type of the event. The convolutional neural network generates an output of representative digital images from the collection based on the importance rating of each digital image.

A method for video frame analysis includes determining a first dissimilarity metric and a second dissimilarity metric. The first dissimilarity metric may correspond to a first difference between a first foreground of a first key frame in a video and a second foreground of a second key frame following the first key frame in the video. The second dissimilarity metric may correspond to a second difference between the second foreground of the second key frame and a third foreground of a third key frame following the second key frame in the video. A playback of the video may be generated based on the first dissimilarity metric and the second dissimilarity metric. Related systems and computer program products are also provided.

Methods and arrangements for automatic ground truth generation of medical image collections. Aspects include receiving a plurality of imaging studies, wherein each imaging study includes one or more images and a textual report associated with the one or more images. Aspects also include selecting a key image from each of the one or more images from each of the plurality of imaging studies and extracting one or more discriminating image features from a region of interest within the key image. Aspects further include processing the textual report associated with the one or more images to detect one or more concept labels, assigning an initial label from the one or more concept labels to the one or more discriminating image features, and learning an association between each of the one or more discriminating image features and the one or more concept labels.

A system and method for filming and discovering videos for users may include receiving an instructional video of an event from a content provider. The instructional video includes portions of a first video stream with images at an eye level of an instructor and portions of a second video stream with close-up shots related to a concept being taught during the event. It is determined whether the instructional video includes a title sequence, an introduction, a lesson, a recap, and a conclusion. If the instructional video fails to contain the title sequence, the introduction, the lesson, the recap, and the conclusion, a rejection of the instructional video is provided that includes an explanation. If the instructional video contains the title sequence, the introduction, the lesson, the recap, and the conclusion, the instructional video is stored in a video database.

Video and corresponding metadata is accessed. Events of interest within the video are identified based on the corresponding metadata, and best scenes are identified based on the identified events of interest. Events of interest can be tagged within the video based on, for instance, user input, audio signals, motion vectors, and metadata corresponding to the video. A camera system can process video data based on the events of interest tagged within the video before outputting the video data. For instance, video scenes associated with tagged events of interest can be combined to form a video highlight clip. Likewise, portions of video tagged with events of interest can be encoded or stored at a higher resolution or frame rate than other portions of the video.