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 #27

Closed HeDo88TH closed 1 year ago

u4gbot commented 1 year ago
Comparison for brighton_beach_small (Differences) Accuracy: 94.11% vs 93.25% | Label | Accuracy | Recall | Precision | F1 | | -------------------- | ---------- | ---------- | ---------- | ---------- | | | 66.95% | 95.07% | 69.36% | 0.80 | | low_vegetation | 21.78% | 48.92% | 28.20% | 0.36 | | | -45.17% | -46.15% | -41.16% | -0.44 | | | | | | | | | 45.35% | 83.98% | 49.64% | 0.62 | | human_made_object | 37.87% | 37.91% | 99.72% | 0.55 | | | -7.48% | -46.08% | 50.08% | -0.07 | | | | | | | | | 14.64% | 99.56% | 14.65% | 0.26 | | building | 19.87% | 45.97% | 25.92% | 0.33 | | | 5.23% | -53.59% | 11.27% | 0.08 | | | | | | | | | 94.06% | 94.12% | 99.93% | 0.97 | | ground | 93.47% | 95.43% | 97.86% | 0.97 | | | -0.59% | 1.31% | -2.08% | -0.00 | | | | | | |
Comparison for sheffield_park_small (Differences) Accuracy: 97.85% vs 23.41% | Label | Accuracy | Recall | Precision | F1 | | -------------------- | ---------- | ---------- | ---------- | ---------- | | | 95.84% | 98.10% | 97.66% | 0.98 | | low_vegetation | 18.98% | 99.96% | 18.98% | 0.32 | | | -76.87% | 1.86% | -78.68% | -0.66 | | | | | | | | | 34.76% | 94.35% | 35.50% | 0.52 | | human_made_object | 1.42% | 1.53% | 16.09% | 0.03 | | | -33.35% | -92.82% | -19.42% | -0.49 | | | | | | | | | 92.46% | 94.00% | 98.27% | 0.96 | | building | 30.28% | 40.01% | 55.44% | 0.46 | | | -62.19% | -53.98% | -42.82% | -0.50 | | | | | | | | | 98.04% | 98.33% | 99.70% | 0.99 | | ground | 3.95% | 4.09% | 54.81% | 0.08 | | | -94.08% | -94.24% | -44.89% | -0.91 | | | | | | |

Average (2 datasets)

Accuracy: 95.98% vs 58.33%

Label Accuracy Recall Precision F1
81.40% 96.58% 83.51% 0.89
low_vegetation 20.38% 74.44% 23.59% 0.34
-61.02% -22.14% -59.92% -0.55
40.06% 89.17% 42.57% 0.57
human_made_object 19.64% 19.72% 57.90% 0.29
-20.41% -69.45% 15.33% -0.28
53.55% 96.78% 56.46% 0.61
building 25.07% 42.99% 40.68% 0.40
-28.48% -53.79% -15.78% -0.21
96.05% 96.22% 99.82% 0.98
ground 48.71% 49.76% 76.33% 0.52
-47.34% -46.47% -23.48% -0.46