OpenDroneMap / ODMSemantic3D

An open photogrammetry dataset of classified 3D point clouds for automated semantic segmentation. CC BY-SA 4.0
Other
15 stars 7 forks source link

Added toledo dataset #23

Closed HeDo88TH closed 1 year ago

u4gbot commented 1 year ago
Comparison for brighton_beach_small (Differences) Accuracy: 94.11% vs 96.13% | Label | Accuracy | Recall | Precision | F1 | | -------------------- | ---------- | ---------- | ---------- | ---------- | | | 66.95% | 95.07% | 69.36% | 0.80 | | low_vegetation | 33.40% | 43.39% | 59.18% | 0.50 | | | -33.55% | -51.67% | -10.18% | -0.30 | | | | | | | | | 94.06% | 94.12% | 99.93% | 0.97 | | ground | 96.42% | 98.61% | 97.75% | 0.98 | | | 2.36% | 4.49% | -2.18% | 0.01 | | | | | | | | | 45.35% | 83.98% | 49.64% | 0.62 | | human_made_object | 39.94% | 40.38% | 97.32% | 0.57 | | | -5.41% | -43.60% | 47.68% | -0.05 | | | | | | | | | 14.64% | 99.56% | 14.65% | 0.26 | | building | 21.57% | 45.97% | 28.89% | 0.35 | | | 6.93% | -53.59% | 14.24% | 0.10 | | | | | | |
Comparison for sheffield_park_small (Differences) Accuracy: 97.85% vs 28.01% | Label | Accuracy | Recall | Precision | F1 | | -------------------- | ---------- | ---------- | ---------- | ---------- | | | 95.84% | 98.10% | 97.66% | 0.98 | | low_vegetation | 19.78% | 99.97% | 19.78% | 0.33 | | | -76.07% | 1.87% | -77.88% | -0.65 | | | | | | | | | 98.04% | 98.33% | 99.70% | 0.99 | | ground | 8.69% | 8.78% | 89.92% | 0.16 | | | -89.34% | -89.55% | -9.79% | -0.83 | | | | | | | | | 34.76% | 94.35% | 35.50% | 0.52 | | human_made_object | 0.83% | 0.85% | 31.14% | 0.02 | | | -33.93% | -93.50% | -4.36% | -0.50 | | | | | | | | | 92.46% | 94.00% | 98.27% | 0.96 | | building | 37.45% | 52.47% | 56.67% | 0.54 | | | -55.01% | -41.53% | -41.59% | -0.42 | | | | | | |

Average (2 datasets)

Accuracy: 95.98% vs 62.07%

Label Accuracy Recall Precision F1
81.40% 96.58% 83.51% 0.89
low_vegetation 26.59% 71.68% 39.48% 0.42
-54.81% -24.90% -44.03% -0.47
96.05% 96.22% 99.82% 0.98
ground 52.56% 53.69% 93.84% 0.57
-43.49% -42.53% -5.98% -0.41
40.06% 89.17% 42.57% 0.57
human_made_object 20.39% 20.62% 64.23% 0.29
-19.67% -68.55% 21.66% -0.28
53.55% 96.78% 56.46% 0.61
building 29.51% 49.22% 42.78% 0.45
-24.04% -47.56% -13.68% -0.16