OverviewCloudSoftwareHardware
OAK Apps
DepthAI
AI
Software Stack
Troubleshooting

ON THIS PAGE

  • CalibrationMetrics
  • Key fields
  • Examples of functionality
  • Reference

CalibrationMetrics

The CalibrationMetrics message is part of the Dynamic Calibration system. It provides compact numeric scores used to evaluate calibration readiness and confidence.It is intended for runtime monitoring and automation logic (for example, deciding when to recalibrate or when to trust the current calibration state).

Key fields

  • dataConfidence: Score between 0 and 1, describing how good the currently collected calibration data is (usefulness of observations).
  • calibrationConfidence: Score between 0 and 1, describing confidence in the current calibration estimate.
Use these metrics together (and compare them against your own baselines/thresholds) when building recalibration policies.

Examples of functionality

CalibrationMetrics is received on metricsOutput when calibration metrics are computed by the Dynamic Calibration node.

Reference

struct

dai::CalibrationMetrics

#include DynamicCalibrationResults.hpp
variable
double calibrationConfidence
variable
double dataConfidence
inline function
CalibrationMetrics()
function
~CalibrationMetrics()
inline function
DatatypeEnum getDatatype()
function
void serialize(std::vector< std::uint8_t > & metadata, DatatypeEnum & datatype)
function

Need assistance?

Head over to Discussion Forum for technical support or any other questions you might have.