Yealink explores Yealink's simplified explanations of common business communication terms regarding the application of three major communication devices, IP Phones, Video Conferencing Equipment and Headsets running on Microsoft, Zoom and other systems, in business work.
When participating in a video conference, it is inevitable that participants will move around, join or leave the conference room. If the camera is fixed and cannot be adjusted, it is difficult to ensure that all participants appear in the video screen, affecting the visual experience; and inefficient manual adjustment of the camera will increase the operational burden.
The Auto Framing function was born, which enables the camera to automatically adjust the angle and magnification according to the number of participants, and output a picture that includes everyone in the meeting room with a suitable size and reasonable layout.
Auto Framing is equipped with a human body detection algorithm. Based on the number of heads identified by the algorithm, the appropriate width-to-height ratio is expanded to serve as the range of the automatic framing area. When the number of people increases/decreases, the algorithm will also calculate a new framing area and adjust the screen ratio adaptively. The zoom lens will continue to track based on the result.
This cycle repeats itself, and Auto Framing ensures that all participants appear in the video screen in full and in the right proportions in real time throughout the meeting.
The Auto Framing function has become a highly popular feature in conference cameras and all-in-one conference systems. In recent years, Yealink has consistently pushed the boundaries of technology, delivering an exceptional experience across all aspects.
Yealink's human body detection algorithm can adjust the torso confidence, which measures the accuracy of detection results for the head, neck, shoulders, chest, waist, and other body parts. When the algorithm identifies a low torso confidence level, it will conduct further object detection to eliminate any misidentified torso, thereby reducing the risk of missed detections or misdetections.
The Yealink experimental team has developed a comprehensive library containing hundreds of models for false detections and missed detections. This library encompasses a wide range of clothing styles, postures, interfering objects, environments, and more. The average recognition accuracy of Yealink's devices is approximately 96%.
Yealink's algorithm employs sophisticated mathematical functions to facilitate smooth movement. When the number of individuals changes, it can select the appropriate image within one second and track movement in real time.
In order to consistently surpass the upper limits of algorithm performance, the strategy employed by the algorithm enhances computational efficiency without extending processing time. This allows Auto Framing to more effectively handle a variety of extreme scenarios.
When there are 30 or more people in the conference room and Auto Framing is continuously activated during a lengthy conference call, the anti-shake performance is maintained within 2.8 milliseconds, ensuring smooth images without lag in extreme scenarios.
Since the algorithm relies on facial recognition, we, as meeting participants, can optimize its performance by minimizing distractions in the meeting room.
> Before using the function, you must calibrate the lens to ensure that no objects are obstructing the fixed-focus lens. Recalibration is necessary after moving the camera.
> Please avoid reflections of individuals, mannequins, cartoon characters, and similar objects in the glass or whiteboard within the meeting room.
> Avoid allowing participants to be in high-exposure areas that could result in missed detections.
MeetingBoard series, MeetingBar AX series, UVC 86, and SmartVision 40 are all equipped with the Auto Framing function.
Related Products
Contact Us
