| 主权项 |
1. A method for presenting metadata during a video conference, comprising:
receiving a first video conferencing stream from a first site; identifying metadata associated with the first video conferencing stream by performing face, voice, or location recognition; creating, using a processor, a second video conferencing stream incorporating the identified metadata, wherein the identified metadata comprises a name of a first person being displayed; presenting the second video conferencing stream at a second site; identifying a topic being discussed in the first video conferencing stream and a third video conferencing stream originating from the second site based at least in part on image recognition performed on the first video conferencing stream and the third video conferencing stream to facilitate identifying the topic being discussed in the first video conferencing stream and the third video conferencing stream; detecting a change that occurs in the first video conferencing stream from the first person being displayed to a second person being displayed; automatically identifying, in a database remote from the second site, supplemental metadata associated with the first video conferencing stream according to the change and according to the identified topic being discussed in the first video conferencing stream and the third video conferencing stream; updating, using the processor, the second video conferencing stream incorporating the identified supplemental metadata, wherein the supplemental metadata comprises a name of the second person being displayed; and presenting the updated second video conferencing stream at the second site. |
