Map the camera

Camera Mapping computes one uniform scale, one rotation, and an X/Y translation from camera pixels to known reference coordinates. A saved mapping can position the overlay when no camera alignment is saved.

This is not lens calibration. It does not remove barrel distortion, fisheye curvature, perspective, or mirroring. Use a fixed camera with a reasonably flat view of the bed.

What you need

An active machine profile and a selected camera.
A current frame captured with Update Overlay.
At least 3 known reference points that are not all on one line.
The image pixel coordinates and reference coordinates for each point.

Steps

1. Prepare reference points

Use points that are easy to identify in the image and whose physical positions you know. Spread them across the useful camera area. A triangle covering much of the bed is the minimum useful layout; 4 or more points provide a better consistency check.

2. Open Camera Mapping

Open the Camera panel.
Click Map Camera.
The Camera Mapping dialog opens with 3 point rows.

3. Enter point pairs

For each row:

Enter Image X / Y, the point's pixel location in the captured frame.
Enter Ref X / Y, the matching physical reference position.
Use Add Point for additional pairs.

All image points must be inside the captured frame. Both the image points and reference points must cover a non-zero area, so 3 points on a straight line cannot be solved.

4. Solve and save

Click Solve. Review the Quality and RMSE values, then click Save Mapping. If you change any point after solving, solve again before saving.

Lower RMSE means the point pairs agree more closely. Do not rely on a fixed Quality threshold by itself. Verify the result against known locations on the bed.

When to redo

You select a different camera.
The camera starts capturing at a different resolution.
You move or rotate the camera mount.
You change the distance between the camera and material surface enough to affect placement.

Map the camera

On this page