Methods and systems for identifying altered content are described herein. The system generates a fingerprint for an unverified content item and locates a plurality of content items that match the fingerprint. The system then compares corresponding frames between the unverified content item and each content item of the plurality of content items. The system identifies, based on the comparing, an altered frame in the unverified content item that does not match a corresponding frame in two or more of the plurality of content items. The system also determines that one or more frames of the unverified content item that follow the altered frame match corresponding frames in the two or more of the plurality of content items. The system then generates for display an indication that the unverified content item contains one or more altered frames.

Systems, methods, and non-transitory computer-readable media can be configured to determine a video embedding for a video content item based at least in part on a first machine learning model. A set of music embeddings can be determined for a set of music content items based at least in part on a second machine learning model. The set of music content items can be ranked based at least in part on the video embedding and the set of music embeddings.

Systems and methods are provided for distributed video storage and search over edge computing devices having a short-term memory and a long-term memory. The method may comprise caching a first portion of data on a first device. The method may further comprise determining, at a second device, whether the first device has the first portion of data. The determining may be based on whether the first piece of data satisfies a specified criterion. The method may further comprise sending the data, or a portion of the data, and/or a representation of the data from the first device to a third device.

A query acquisition unit (610) acquires a query video. A query frame selection unit (620) selects a plurality of query frames from among a plurality of first frame images included in the query video. A video selection unit (630) selects a video by using the plurality of query frames. At this time, the video selection unit (630) uses references (1) and (2) below as a condition for selecting the video.

Reference (1) A condition that a “similar frame image whose degree of similarity to a query frame satisfies a first reference is present” is satisfied for at least two query frames.

Reference (2) When query frames associated with each of at least two similar frame images are arranged in the same order as that of the at least two similar frame images, the arrangement order coincides with an arrangement order of at least two query frames in a query video.

Disclosed herein is an image analysis system including an information extraction section, a correct answer generation section, and a training section. The information extraction section extracts an input region and auxiliary information from video content, the input region being a portion of each of multiple images constituting the video content, the auxiliary information being information different from the input region. The correct answer generation section generates correct data from the extracted auxiliary information. The training section trains a machine learning model by use of training data including the input region and the correct data.

A method and system for identifying social trends are provided. The method includes collecting multimedia content from a plurality of data sources; gathering environmental variables related to the collected multimedia content; extracting visual elements from the collected multimedia content; generating at least one signature for each extracted visual element; generating at least one cluster of visual elements by clustering at least similar signatures generated for the extracted visual elements; correlating environmental variables related to visual elements in the at least one cluster; determining at least one social trend by associating the correlated environmental variables with the at least one cluster.

A server includes a processor programmed to: acquire first metadata of a first media file recorded by a first mobile device; acquire second metadata of a second media file recorded by a second mobile device; determine that the first media file and the second medial file are likely recordings of the same event when a similarity exceeds a first threshold. The processor is further programmed to, when the first media file and the second medial file are likely recordings of the same event: determine, based on a comparison between the first media file and the second media file, which of the first media file and the second media file is a higher quality recording of the same event; and when the first media file is the higher quality recording, send a link to the first media file to the second mobile device.

Systems and methods for replacing original media bookmarks of at least a portion of a digital media file with replacement bookmarks is described. A media fingerprint engine detects the location of the original fingerprints associated with the portion of the digital media file and a region analysis algorithm characterizes regions of media file spanning the location of the original bookmarks by data class types. The replacement bookmarks are associated with the data class types and are overwritten or otherwise are substituted for the original bookmarks. The replacement bookmarks then are subjected to a fingerprint matching algorithm that incorporates media timeline and media related metadata.

A system includes a computing platform having processing hardware and a memory storing a software code. The processing hardware is configured to execute the software code to receive multiple content units each including a start descriptor for an initial content segment and an end descriptor for a last content segment, identify the start descriptor and the end descriptor for each of the content units, and select a first content unit for beginning a content compilation. The processing hardware is further configured to execute the software code to determine multiple similarity metrics each comparing the end descriptor of the first content unit with the start descriptor of a respective one of the other content units, rank, using the similarity metrics, the other content units with respect to one another, select, based on the rank, a second content unit, and composite the content compilation using the second content unit.

Videos may be automatically generated using a set of video clip. Individual moments of interest may be identified within individual video clips of a set of video clips. A moment of interest may correspond to a point in time within a video clip. The point in time may be associated with one or more values of one or more attributes of the video clip. Individual moments of interest may be associated with individual portions of a video. The video may be generated using the set of video clips based on the associations.