- Number
- 20220327402
- Published
- 2022-10-13
- Filed
- 2021-04-08
- Assignee
- Disney Enterprises, Inc.
- Inventors
- Tiwari; Sanchita, Van Dall; Dirk, Devassy; Jayadas
- CPC
- G06F18/22; G06N20/00; G06N5/04; G06V10/74; G06V10/751; G06V20/46; G06V20/48; G11B27/031; G11B27/28
- Verdict
- Set aside content compilation tool, business
- Source
- Google Patents · FreePatentsOnline
Abstract
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.
Background
BACKGROUND
Joining or merging discrete units of content is a common task in content creation. For example, multiple video clips may be merged by a professional video editor in a studio, or by an amateur user via an online social media platform or utilizing a commercial software application. However, conventional solutions for merging discrete content units to form a content compilation achieve good results only when working with content that has a common theme or context running through the content units. For instance, where the content units are video clips, the clips being merged are typically from sporting events, celebrations, vacations, home videos, or educational tutorials, to name a few examples. In the conventional art, video clips selected from multiple different topics or themes tend to produce poor composite videos that may be jarring to watch and listen to due to abrupt transitions of visual style, thematic subject matter, and sound effects. Consequently, there is a need in the art for an automated solution for compositing disparate units of content from a broad range of topics, many of which may have different audio-visual characteristics or styles, to produce a compilation of that content that is appealing, especially where these compilations are unique for each user and human involvement cannot be afforded.
Claims
1. A system comprising: a computing platform including a processing hardware and a system memory storing a software code; the processing hardware configured to execute the software code to: receive a plurality of content units, each one of the plurality of content units 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 one of the plurality of content units; select a first content unit of the plurality of content units for beginning a content compilation; determine a plurality of similarity metrics each comparing the end descriptor of the first content unit with the start descriptor of a respective one of other content units of the plurality of content units; rank, using the plurality of similarity metrics, the one other content unit of the plurality of content units with respect to one another; select, based on the rank, a second content unit of the plurality of content units; and composite the content compilation using the second content unit. ||
9. A method for use by a system including a computing platform having a processing hardware, and a system memory storing a software code, the method comprising: receiving, by the software code executed by the processing hardware, a plurality of content units, each one of the plurality of content units including a start descriptor for an initial content segment and an end descriptor for a last content segment; identifying, by the software code executed by the processing hardware, the start descriptor and the end descriptor for each one of the plurality of content units; selecting, by the software code executed by the processing hardware, a first content unit of the plurality of content units for beginning a content compilation; determining, by the software code executed by the processing hardware, a plurality of similarity metrics each comparing the end descriptor of the first content unit with the start descriptor of a respective one of other content units of the plurality of content units; ranking, by the software code executed by the processing hardware and using the plurality of similarity metrics, the one other content unit of the plurality of content units with respect to one another; selecting, by the software code executed by the processing hardware and based on the ranking, a second content unit of the plurality of content units; and compositing, by the software code executed by the processing hardware, the content compilation using the second content unit. ||
17. A system comprising: a computing platform including a processing hardware and a system memory storing a software code; the processing hardware configured to execute the software code to: receive a plurality of content units, each one of the plurality of content units 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 one of the plurality of content units; identify a plurality of candidate content compilations, each of the plurality of candidate compilations including the plurality of content units in a different temporal sequence; for each one of the plurality of candidate content compilations, determine, by comparing the end descriptor of each of the plurality of content units to the start descriptor of a next one of the plurality of content units in the different temporal sequence, a score for the one of the plurality of candidate content compilations, resulting in a plurality of scores; and select, based on the plurality of scores, a best content compilation from among the plurality of candidate content compilations.