Filtering video for video telephony and video conferencing
US9253443B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Mar 3, 2015 |
| Grant date | Feb 2, 2016 |
| Priority date | — |
| Expiry date | Mar 3, 2035 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG10L15/26
- WIPO fieldAudio-visual technology
- WIPO sectorElectrical engineering
Abstract
In an approach to video filtering, a computer receives a first video frame of a presenter from a video feed that includes an audio feed. The computer extracts a face of the presenter and a background from the first video frame. The computer receives a second video frame. The computer determines whether a difference between the first video frame and the second video frame exceeds a pre-defined threshold. The computer converts the audio feed from speech to text. The computer determines whether the difference between the first video frame and the second video frame is expected, based, at least in part, on the converted audio feeds. The computer discards video data associated with the difference between the first video frame and the second video frame. The computer creates a third video frame, based, at least in part, on non-discarded video data.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.