Demo | VL3D++

Home

RGB view—Point cloud in real photographic colour

vsFinal result

Loading visualisation…

◀▶

RGB view

Final result

⇔Drag to compare

Data provider

Canteras de Santullán S.A.

Summary

95.1%

Overall Accuracy

92.6%

Precision

96.4%

Recall

94.3%

F1 Score

Full metrics

IoU?

Intersection over Union (macro)Overlap between prediction and ground truth per class, averaged unweighted. Reference values are lower in 3D point clouds than in 2D imagery.0–100% · >40% acceptable · >60% good in 3D

89.3%

wP?

Weighted PrecisionPer-class precision weighted by support (number of points). Reflects practical performance weighted by the real dataset distribution.0–100%

95.4%

wR?

Weighted RecallPer-class recall weighted by support. In point-by-point classification with mutually exclusive classes, its numerical value equals Overall Accuracy.0–100% · Expected value ≈ OA

96.4%

wF1?

Weighted F1 ScoreHarmonic mean of wPrecision and wRecall, weighted by class support. More informative than OA with imbalanced classes.0–100%

94.3%

wIoU?

Weighted IoUPer-class IoU weighted by support. Gives more weight to larger classes. Macro IoU is more sensitive to minority-class performance.0–100%

90.8%

MCC?

Matthews Correlation CoefficientCorrelation between predictions and ground truth. Considers all confusion-matrix elements. Reliable with imbalanced classes.-100 to 100 · near 0 = no correlation · 100 = perfect

93.3%

Kappa?

Cohen's KappaAgreement between model and ground truth corrected for chance agreement. More robust than OA on imbalanced datasets.0–100% · <40% poor · 60–80% good · >80% very good

93.3%

Macro vs Weighted

Metric

Macro

Weighted

Precision

92.6%

95.4%

Recall

96.4%

94.3%

IoU

89.3%

90.8%

Macro = unweighted arithmetic mean per class. More sensitive to minority-class performance. Weighted = weighted by class support (number of points). Better reflects practical impact on the real dataset.

Layer guide

RGB view
Point cloud coloured with its real RGB values (photogrammetry or LiDAR-co-registered camera). Serves as visual context before the model analysis.

Raw point cloud
3D point cloud without labels. Colour encodes LiDAR return intensity or normalised elevation.

Error mask
Per-point classification accuracy vs ground truth. Red = likely error. White/grey = correct classification.

Uncertainty
Entropy of the predicted class probability distribution per point. Yellow = maximum doubt. Purple = high confidence.

Ground truth
Expert-validated semantic labelling. Reference against which all model metrics are computed.

Final result
Model prediction over the entire cloud. Each point assigned its semantic class. Main VL3D++ output.

RGB view—Point cloud in real photographic colour

vsFinal result

Loading visualisation…

◀▶

RGB view

Final result

⇔Drag to compare

Data provider

Canteras de Santullán S.A.

Summary

95.1%

Overall Accuracy

92.6%

Precision

96.4%

Recall

94.3%

F1 Score

Full metrics

IoU?

89.3%

wP?

Weighted PrecisionPer-class precision weighted by support (number of points). Reflects practical performance weighted by the real dataset distribution.0–100%

95.4%

wR?

96.4%

wF1?

Weighted F1 ScoreHarmonic mean of wPrecision and wRecall, weighted by class support. More informative than OA with imbalanced classes.0–100%

94.3%

wIoU?

Weighted IoUPer-class IoU weighted by support. Gives more weight to larger classes. Macro IoU is more sensitive to minority-class performance.0–100%

90.8%

MCC?

93.3%

Kappa?

Cohen's KappaAgreement between model and ground truth corrected for chance agreement. More robust than OA on imbalanced datasets.0–100% · <40% poor · 60–80% good · >80% very good

93.3%

Macro vs Weighted

Metric

Macro

Weighted

Precision

92.6%

95.4%

Recall

96.4%

94.3%

IoU

89.3%

90.8%

Layer guide

RGB view
Point cloud coloured with its real RGB values (photogrammetry or LiDAR-co-registered camera). Serves as visual context before the model analysis.

Raw point cloud
3D point cloud without labels. Colour encodes LiDAR return intensity or normalised elevation.

Error mask
Per-point classification accuracy vs ground truth. Red = likely error. White/grey = correct classification.

Uncertainty
Entropy of the predicted class probability distribution per point. Yellow = maximum doubt. Purple = high confidence.

Ground truth
Expert-validated semantic labelling. Reference against which all model metrics are computed.

Final result
Model prediction over the entire cloud. Each point assigned its semantic class. Main VL3D++ output.