| CPC G06T 7/12 (2017.01) [G06T 7/194 (2017.01); G06T 7/90 (2017.01); G06T 11/40 (2013.01); G06V 10/25 (2022.01); G10L 25/78 (2013.01); H04R 3/005 (2013.01)] | 16 Claims |

|
9. A panoramic video conference method, being adapted for use in a panoramic video conference system, the panoramic video conference system comprising a panoramic video generating apparatus and a control apparatus, and the panoramic video conference method is performed by the control apparatus and comprises following steps:
analyzing a panoramic video received from the panoramic video generating apparatus to identify a plurality of video objects in the panoramic video;
selecting a video object to be removed from the video objects based on a privacy mode;
removing a video corresponding to the video object to generate a blank area;
generating a background filling video according to a plurality of adjacent blocks of the video object being removed; and
filling the background filling video to the blank area to generate a panoramic conference video;
wherein the panoramic video conference system further comprises:
a microphone array, being connected to the control apparatus through a network connection, and being configured to sense a sound of a conference room to generate a sound signal; and
a display, being configured to play a remote conference video and a remote conference sound;
wherein the panoramic video conference method further comprises following steps:
receiving the sound signal from the microphone array;
determining a speech source based on the sound signal;
comparing a plurality of conference member features with the panoramic video corresponding to the speech source to determine whether the panoramic video corresponding to the speech source has a first conference member feature that matches the conference member features;
generating a focusing video corresponding to the speech source when it is determined that the panoramic video corresponding to the speech source has the first conference member feature;
generating the panoramic conference video based on the focusing video and the panoramic video;
generating an exclusion area based on a position of the display;
comparing the speech source with the exclusion area to determine whether the speech source comes from the exclusion area; and
wherein the focusing video corresponding to the speech source is not generated when the speech source comes from the exclusion area.
|