Repeatability and QA

Cross-cutting topic: making Metashape projects reproducible and auditable — same dataset, same parameters, same numbers — and extracting from them the metrics a quality-assurance process can gate on.

Articles

• Reproducing chunk-info statistics in Python
• Keypoint-size-normalised reprojection error: the kps metric
• Tie-point multiplicity: track length, distribution, and what it tells you
• Reprojection error analysis: per-camera and per-tie-point
• Marker projection statistics: counts, per-marker errors, metres vs pixels
• Scalebar distance error: per-scalebar values and RMS aggregation
• Camera reference error: per-camera location & orientation in Python
• Sensor and camera shared-tie-point graphs: detecting isolated groups
• DSM ridge-line artefacts: alignment-quality diagnosis
• Saving estimated reference values to file: location, rotation, error, and sigma

Coverage

This section reproduces every chunk-info statistic the GUI displays plus several derived QA metrics that aren't shown directly. Together they let a Python script reproduce the Survey Data / Cameras / Reference pages of a Metashape PDF report and compute additional quality gates beyond it.