We want to know how the algorithm of the EagleEye Director will work with a speaker that is moving in a room during a speech.
Watching the demo videos, we know at the beginning the product will switch from a general view (Cam1) to a zoomed view of the speaker (Cam2). However, what will happen when the speaker, slowly, starts to exit from the camera view?
1. The active and zoomed camera (Cam2) will slowly move (pan) to follow the speaker (That is the only souce of sound in the room)
2. When the speaker moves and he/she is out of the view, the Director will not track (panning). It will switch again to the general view camera (Cam1) and some seconds after, it will return again to the zoomed view of the speaker (Cam2) in their new location in the room.
It is important to clarify that the speaker will not move very fast, just walking during the speech.
If you have some demo video explaining this situation, it will be great.
The technology used in the EagleEye Director is called ‘dynamic speaker locating technology’, it is not tracking technology. The EagleEye Director will locate and intelligently frame the speaker within a frame, the speaker cannot be moving. The framing will happen first with local audio, to find the person within the room, then will frame the person when it can identify the eyes, nose and mouth. The eyes will be at the 1/3 way down the frame.
In contrast the EagleEye Producer is ‘dynamic room framing technology’. The framing camera, part of the EagleEye Producer device, monitors the room for participants and then provides the attached EagleEye camera coordinates to use. The technology will recognize between portraits (they do not move) and people (they move) within the room.
Thank you very much for your answer!
If I understood correctly, when the speaker starts to move and half face will be out of the frame, the Director will switch to the General View (Cam1) and some seconds later the speaker will appear again zoomed in the new location (Cam 2). Is that correct?
If someone moves the Director will go the establishing shot. When the speaker is in one place, i.e. sitting down at a table or behind a podium, if the Director can identify the eyes, nose and mouth it will then switch to the speaker (immersive shot).