System and method for augmented reality video conferencing
A system includes a plurality of capturing devices and a plurality of displaying devices. The capturing devices and the displaying devices can be communicatively connected to a server. The server can receive captured data from the capturing devices and transform the data into a digitized format. At least one of the capturing devices can record a video during a meeting and the capturing device can transmit the captured video to the server as a video feed. The video feed can show an area that includes handwritten text, e.g., a whiteboard. The server can receive the video feed from the capturing device and perform various processes on the video. For example, the server can perform a voice recognition, text recognition, handwriting recognition, face recognition and/or object recognition technique on the captured video.
1. A method comprising:
receiving, at a server, a first video feed from a first camera;
processing, using a processor of the server, the first video feed to:
extract a first text from a first segment of the first video feed; and
identify a first person;
receiving, at the server, a second video feed from a second camera;
processing, using the processor of the server, the second video feed to:
extract a second text from a second segment of the second video feed; and
identify a second person;
generating, using the processor of the server, a slide showing a first summary of the first text and a second summary of the second text, wherein:
the first summary is displayed in association with the first person and the second summary is displayed in association with the second person; and
the first summary is selectable to display a first media file and the second summary is selectable to display a second media file.
2. The method of claim 1 , further comprising processing, using the processor of the server, the first video feed to extract a first shape from the first segment of the first video feed.
3. The method of claim 2 , wherein the processor extracts the first shape using a shape recognition module.
4. The method of claim 2 , wherein the slide is generated to show the first shape in digital format.
5. The method of claim 1 , wherein the slide further displays the first text or the second text.
6. The method of claim 5 , wherein the processor generates the first summary using a machine learning technique.
7. The method of claim 5 , wherein the processor generates the first summary using a natural language processing technique.
8. The method of claim 1 , wherein the processor identifies the first person using a face recognition technique.
9. The method of claim 1 , wherein the processor identifies the first person using a voice recognition technique.
10. The method of claim 1 , wherein the first media file is the first segment of the first video feed.
11. The method of claim 1 , wherein the first media file is a voice recording.
12. The method of claim 1 , wherein the second media file is the second segment of the second video feed.
13. The method of claim 1 , wherein the first summary is selectable to display an identity of the first person.
14. The method of claim 1 , wherein the slide is included in a PDF document or a Word document.
15. The method of claim 1 , wherein the slide is text-searchable.
16. The method of claim 1 , wherein the processor extracts the first text from the first segment of the first video feed using a transcription module.
17. The method of claim 1 , wherein the processor extracts the first text from the first segment of the first video feed using a text recognition module.
18. The method of claim 1 , wherein the processor extracts the first text from the first segment of the first video feed using a handwriting recognition module.
19. The method of claim 1 , wherein the processor identifies the first person using a facial recognition module configured to:
scan the first video feed;
detect a face in the first video feed; and
identify the face based on a comparison of the face with a plurality of photos of faces stored in a database.
20. A method comprising:
receiving, at a server, a first video feed from a first camera;
processing, using a processor of the server, the first video feed to:
subtract a background from the first video feed to generate an overlay feed; and
identify a first person;
receiving, at the server, a second video feed from a second camera of an AR glass;
processing, using the processor of the server, the second video feed to:
extract a second text from a second segment of the second video feed; and
identify a second person; and
transmit, using the processor of the server, the overlay feed to the AR glass, wherein the overlay feed is configured to be superimposed over a field of view of the AR glass.