(Camera)¶
The monocular camera’s perspective to render from.
If this parameter is given, following parameters will be ignored:
(Size): The size of the rendered point map will match the size of the monocular camera image.
(PixelSize): To render the meshes, the camera’s internal projection and distortion parameters will be used instead.
(ViewPose): The monocular camera’s world pose will be used instead.
After the command execution, each pixel in RenderPointMap holds three float values, corresponding to the x, y and z world coordinates of the surfaces seen by the monocular camera at the same pixel coordinate. Thus, you can directly overlay the two images to obtain a 6 channel (R, G, B, X, Y, Z) image (or 4 channels (Gray, X, Y, Z) for a black and white viewer camera).
Note
When using perspective rendering, you need to specify the (Near) and (Far) parameters explicitly, because both clipping planes are required to be in front of the camera, i.e., positive! You can, for example, set the (Near) clipping to 50, as the default value for (Far) is already positive. Beware that this setting will discard all surfaces closer than 50 mm to the monocular camera from the rendered image, so you must carefully choose the two planes to not clip away useful depth data!
JSON Format¶
- Type
Any of the types listed below.
- Default Value
All open cameras in the current NxLib instance.
Types¶
Type |
Description |
|---|---|
|
A single serial number of a camera. |
|
A single EEPROM ID of a camera. |
|
An array containing a single serial number or EEPROM ID as specified above. |