| Keterangan Lainnya |
Using 8-32mm lens to meet the recording and broadcasting of large classrooms
1. Support for outputting panoramic images, the resolution of panoramic images can reach 2 million pixels (1920 × 1080), and it can output 30fps real-time image at this resolution
2. Support the output of close-up pictures, the resolution of close-up pictures can reach 2 million pixels (1920 × 1080), and can output 30fps real-time image at this resolution
3. Using deep learning algorithms, taking massive pictures and video resources as the roadbed, and extracting target features by the machine itself, forming a deep learning human body image. Greatly improve the detection rate of the target human body.
4. Support the output of teacher close-up pictures, teacher panoramic pictures, and blackboard close-up pictures, and the resolution of each picture is not less than 1080p
5. Support 2 kinds of teacher close-up tracking mode switching: real-time tracking mode and dynamic tracking mode
6. Real-time tracking: The built-in tracking system uses image recognition technology to accurately locate the teacher's location, real-time tracking and shooting, and can switch the close-up mode, the tracking process is smooth and smooth, and the visual experience is comfortable
7. Dynamic tracking: Different from the electronic pan-tilt to track the teacher in real time, it continuously tracks the teacher's human body problem center, and adopts the switching state of the teacher standing close-up and walking around the panorama to ensure that the screen switching is reasonable and not dazzling
8. In the recording mode, support does not need to configure other analysis servers to achieve single-camera face comparison, and achieve teacher attendance |
| Deskripsi |
Using 8-32mm lens to meet the recording and broadcasting of large classrooms
1. Support for outputting panoramic images, the resolution of panoramic images can reach 2 million pixels (1920 × 1080), and it can output 30fps real-time image at this resolution
2. Support the output of close-up pictures, the resolution of close-up pictures can reach 2 million pixels (1920 × 1080), and can output 30fps real-time image at this resolution
3. Using deep learning algorithms, taking massive pictures and video resources as the roadbed, and extracting target features by the machine itself, forming a deep learning human body image. Greatly improve the detection rate of the target human body.
4. Support the output of teacher close-up pictures, teacher panoramic pictures, and blackboard close-up pictures, and the resolution of each picture is not less than 1080p
5. Support 2 kinds of teacher close-up tracking mode switching: real-time tracking mode and dynamic tracking mode
6. Real-time tracking: The built-in tracking system uses image recognition technology to accurately locate the teacher's location, real-time tracking and shooting, and can switch the close-up mode, the tracking process is smooth and smooth, and the visual experience is comfortable
7. Dynamic tracking: Different from the electronic pan-tilt to track the teacher in real time, it continuously tracks the teacher's human body problem center, and adopts the switching state of the teacher standing close-up and walking around the panorama to ensure that the screen switching is reasonable and not dazzling
8. In the recording mode, support does not need to configure other analysis servers to achieve single-camera face comparison, and achieve teacher attendance |
