Conference scene picture self-adaption method

Abstract

The invention discloses a conference scene picture self-adaption method. The method comprises the steps of: S1, enabling a camera to collect a large-view image of a video conference; S2, performing image analysis on the large visual field, and calculating positions and coordinate information of participants in the large visual field image; S31, in the mechanical pan-tilt, calculating a position where the mechanical pan-tilt needs to drive the camera to move and a position where a camera lens needs to zoom, enabling participants to be in a video range, and taking the video picture containing all participants as an output video image; and S32, in the electronic pan-tilt, cutting an image containing all participants based on the large-view image, and taking the image as an output video image.An image analysis mode is adopted, personnel position distribution and coordinates in the conference room are dynamically calculated, dynamic self-adaptive adjustment is achieved, misoperation of manual control is avoided, conference video scene self-adaption is automatically and dynamically achieved, and a better video is presented to conference participants of the opposite side.

Description

technical field [0001] The invention relates to the field of video picture self-adaptation, in particular to a method for meeting scene picture self-adaptation. Background technique [0002] In the existing technology, with the rapid development of computer network and video technology, the communication between people and multiple people is no longer limited by different regions. Remote video conferencing has become an important way of communication. The video of the conference room can be connected to the same conference system to conduct multi-party video conferences. [0003] Since the participants are all through the audio and video system, a conference camera that can cover the entire conference room will be installed in the general conference room, which can ensure that all participants are in the video field of view, thereby enhancing the face-to-face feeling of the conference. In order to ensure that all participants are in the video field of view, and ensure that ...

AI Technical Summary

Problems solved by technology

In order to ensure that all participants are in the video field of view, and ensure that the participants occupy a large proportion in the video, it is realized by manually controlling the meeting camera pan / tilt. Sometimes the meeting personnel move away or There are new participants joining, which brings great inconvenience, and there will be misoperation and poor real-time performance

Images