| 摘要 |
A system for generating video images and corresponding audio of multiple parties engaged in a video conference is provided. The system includes multiple voice transducers for receiving voice signals from the multiple parties and a video camera for capturing moving images of one of the multiple parties who is speaking. The system further includes a processor in communication with the voice transducers and video camera. The processor determines respective distances and angles between the party who is speaking and each of the multiple voice transducers. Additionally, the processor identifies a location of the party who is speaking relative to the video camera based on the determined respective distances and angles. Based on the identified location, the processor selects a portion of a video frame produced by video camera and processes the selected portion to mitigate optical distortion and generate an overall picture in which the party speaking does not appear unnaturally small relative to the overall picture.
|
