Video streaming services insert in-stream video ads at various points when providing users video content. However, such ad insertion can sometimes affect the viewing experience of the user if the content is paused at inappropriate moments or by unnecessarily prolonging the overall viewing session. This disclosure describes techniques to analyze video content using machine learning models or other suitable techniques to identify suitable portions that can be replaced by inserted ads or can be displayed together with an ad. By replacing sections of the video with little or no content, the techniques provide a more integrated, resource-efficient, and generally improved ad experience for users. For example, the user experience for video content such as cooking videos, exercise videos, or other videos with natural pauses can be improved by such analysis and ad insertion.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Carbune, Victor and Sharifi, Matthew, "Identifying Suitable Slots for In-stream Video Advertisements", Technical Disclosure Commons, (July 07, 2021)