Adjusting the FOV on the RealSense D457
Is there any way to adjust the FOV of D457? If there is, please let me know. And I want to check the RMS error, how do I set the subpixel of D457?
-
Hi Jhheo The field of view of the image will vary depending on which resolution is selected. The smaller the resolution, the more restricted the view of the scene will be. So you can affect the image's field of view size through the choice of resolution, but not custom-design the size.
If you asking about subpixel linearity, also known as the A-factor, then Intel's guide to the subject at the link below may be helpful.
In regard to setting the A-factor value, there is a Python script reference here:
-
I'm grateful for your help every time. I have an additional issue. The accuracy of the depth data is very poor. When measuring a desk and a chair about 2 meters away, the depth data shows them to be between 5 to 15 meters apart. The actual distance difference between the desk and chair is less than 1 meter. What could be the reason for this discrepancy, and are there any solutions?
-
Are the desk or chair dark grey or black in color? If they are then this can cause depth to be missing or highly inaccurate. This is because it is a general physics principle (not specific to RealSense) that dark grey or black absorbs light and so makes it more difficult for depth cameras to read depth information from such surfaces. The darker the color shade, the more light that is absorbed and so the less depth detail that the camera can obtain. Projecting a strong light source onto such surfaces can help to bring out depth detail.
-
The reason why the chair would be inaccurate is clear then. Plain white surfaces (such as white walls) can also be difficult for the camera to analyze. You could try increasing the Laser Power option to its maximum value of '360' to see whether it improves results by making the projected infrared dot pattern that is cast onto surfaces in the scene more visible to the camera.
-
If the camera is observing asphalt in a sewer and the asphalt is dark grey / black then the results may not be as bad as observing a smooth surface like a chair / desk. Asphalt will have texture on its surface, which helps the camera with depth analysis, and the infrared dot pattern projection is more strongly visible to the camera with some types of material than other types.
A YouTube video at the link below of a RealSense-equipped car travelling fast on a road may give a useful demonstration of what depth results can be expected from asphalt type material.
Please sign in to leave a comment.
Comments
7 comments