Requirements for the operation of the Human pose detector and its sub-detectors are given in the table:
The Pose detection tools operation has the following requirements:
...
...
- We also don't recommend using a resolution higher than
|
...
- 1920×1080, since higher resolution
|
...
- doesn't increase the detection quality, but significantly increases the consumption of resources. The optimal resolution for solving typical tasks is 1280x720
|
...
- The frame rate per second in the video stream from the camera is at least 8
|
...
...
| Note |
|---|
| - Correct operation of the detector isn't guaranteed if you use frames that were masked before they got into Axxon One by an IP camera or similar methods.
- Correct operation of the Human pose detector sub-detectors isn't guaranteed if you use fisheye cameras and other cameras with strong image distortion.
- The camera tilt angle from top to bottom on a person must not exceed 30°. Otherwise, the detector won't operate correctly.
The following diagrams show the examples of:- The acceptable tilt angle:
 Image Added - The unacceptable tilt angle:
 Image Added
|
|
| Lighting | |
...
- is at least 50 lux per square meter. In conditions of insufficient or excessive lighting (night or light-striking), stable operation of the analytics
|
...
...
...
- No abrupt changes in lighting
|
...
...
...
| - Moving objects are visually separable from each other
|
...
- The background is mostly static and
|
...
...
- Moving objects are minimally obscured by
|
...
- static objects in the scene (columns, trees,
|
...
...
...
- doesn't have reflective surfaces and sharp shadows from moving objects. If present, they
|
...
...
...
- doesn't result in image offsets greater than 1% of the frame size
|
...
| Note |
|---|
|
Correct operation of the Pose detection tools is not guaranteed when using a fish-eye lens. |
- For the correct operation of the Sitting person detector and Person down detector sub-detectors, the area of interest must be covered from at least two perpendicular angles (a and b are the directional beams from the cameras and the point of their intersection is the area of interest):
 Image Added
The algorithms have a limitation and can work incorrectly if:
- a person falls down or sits down with their head or face towards the camera,
- a person falls down with their feet towards the camera (for the Person down detector sub-detector)
|
...
Hardware requirements for neural analytics operation
...
