Evaluation metrics for 3D scene graph generation.
| Metric | Definition | Use Case |
|---|---|---|
| R@K | Fraction of GT triplets recovered in top-K | Overall relation recall |
| mR@K | Avg. Recall@K over all predicate classes | Mitigates long-tail issue |
| ZS R@K | Recall@K on unseen subject-predicate-object combinations | Tests generalization to novel relations |
| F1 Score | Harmonic mean of precision and recall | Balances precision and recall |
| mAP | Area under precision-recall curve | Standard detection metric |
| Metric | Definition | Use Case |
|---|---|---|
| R@K | Fraction of GT triplets recovered in top-K | Overall relation recall |
| mR@K | Avg. Recall@K over all predicate classes | Mitigates long-tail issue |
| ZS R@K | Recall@K on unseen subject-predicate-object combinations | Tests generalization to novel relations |
| F1 Score | Harmonic mean of precision and recall | Balances precision and recall |
| mAP | Area under precision-recall curve | Standard detection metric |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.