DEPTH_PRECISION Class Reference

Lists the precisions available for neural depth inference. More...

Static Functions

def from_string (str, name)
 Converts a string to the corresponding DEPTH_PRECISION enum value. More...
 

Detailed Description

Lists the precisions available for neural depth inference.

Enumerator
FP16 Half-precision neural depth inference (default).
INT8 INT8 neural depth inference: faster and with a lower memory footprint, for a small accuracy cost.
Only sl.DEPTH_MODE.NEURAL provides an INT8 model, the other depth modes always run in FP16.
Always safe to set: the SDK falls back to FP16, and says so in the log, when the depth mode or the GPU does not support INT8.
LAST Neither of the above.
Only ever returned by Camera.get_init_parameters(), never set by the user: the depth engine runs at a precision that has no sl.DEPTH_PRECISION value.

Functions

◆ from_string()

def from_string (   str,
  name 
)
static

Converts a string to the corresponding DEPTH_PRECISION enum value.

Parameters
name: The string representation of the DEPTH_PRECISION enum value.
Returns
The corresponding DEPTH_PRECISION enum value. If the string does not match any DEPTH_PRECISION, DEPTH_PRECISION.LAST is returned.