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

Add waterbury point cloud sampled and mapping files #41

Closed HeDo88TH closed 1 year ago

HeDo88TH commented 1 year ago

As title says!

u4gbot commented 1 year ago
brighton_beach_small Overall accuracy from 63.73% to 70.54% | human_made_object | Accuracy | Recall | Precision | F1 | | -------------------- | -------- | -------- | --------- | -------- | | Before | 0.00% | 0.00% | N/A | N/A | | After | 36.77% | 71.48% | 43.10% | 0.54 | | Diff | N/A | N/A | N/A | N/A | | ground | Accuracy | Recall | Precision | F1 | | -------------------- | -------- | -------- | --------- | -------- | | Before | 63.73% | 64.12% | 99.06% | 0.78 | | After | 70.39% | 71.09% | 98.64% | 0.83 | | Diff | 6.67% | 6.97% | -0.42% | 0.05 | | building | Accuracy | Recall | Precision | F1 | | -------------------- | -------- | -------- | --------- | -------- | | Before | 1.87% | 99.96% | 1.87% | 0.04 | | After | 1.43% | 59.51% | 1.44% | 0.03 | | Diff | -0.44% | -40.46% | -0.43% | -0.01 | | low_vegetation | Accuracy | Recall | Precision | F1 | | -------------------- | -------- | -------- | --------- | -------- | | Before | 55.99% | 58.07% | 94.01% | 0.72 | | After | 51.91% | 56.02% | 87.60% | 0.68 | | Diff | -4.09% | -2.05% | -6.40% | -0.03 |
sheffield_park_small Overall accuracy from 91.37% to 80.31% | human_made_object | Accuracy | Recall | Precision | F1 | | -------------------- | -------- | -------- | --------- | -------- | | Before | 12.76% | 86.17% | 13.03% | 0.23 | | After | 4.50% | 95.31% | 4.51% | 0.09 | | Diff | -8.27% | 9.15% | -8.52% | -0.14 | | ground | Accuracy | Recall | Precision | F1 | | -------------------- | -------- | -------- | --------- | -------- | | Before | 91.62% | 92.53% | 98.94% | 0.96 | | After | 85.57% | 86.11% | 99.27% | 0.92 | | Diff | -6.05% | -6.41% | 0.33% | -0.03 | | building | Accuracy | Recall | Precision | F1 | | -------------------- | -------- | -------- | --------- | -------- | | Before | 75.33% | 76.46% | 98.08% | 0.86 | | After | 6.80% | 6.81% | 98.25% | 0.13 | | Diff | -68.52% | -69.65% | 0.17% | -0.73 | | low_vegetation | Accuracy | Recall | Precision | F1 | | -------------------- | -------- | -------- | --------- | -------- | | Before | 80.03% | 94.96% | 83.58% | 0.89 | | After | 81.77% | 95.67% | 84.92% | 0.90 | | Diff | 1.74% | 0.71% | 1.34% | 0.01 |

Average (2 datasets)

Overall accuracy from 77.55% to 75.43%

human_made_object Accuracy Recall Precision F1
Before 6.38% 43.08% 6.51% 0.11
After 20.63% 83.40% 23.80% 0.31
Diff 14.25% 40.31% 17.29% 0.20
ground Accuracy Recall Precision F1
Before 77.67% 78.32% 99.00% 0.87
After 77.98% 78.60% 98.95% 0.87
Diff 0.31% 0.28% -0.05% 0.01
building Accuracy Recall Precision F1
Before 38.60% 88.21% 49.97% 0.45
After 4.11% 33.16% 49.84% 0.08
Diff -34.48% -55.05% -0.13% -0.37
low_vegetation Accuracy Recall Precision F1
Before 68.01% 76.51% 88.79% 0.80
After 66.84% 75.84% 86.26% 0.79
Diff -1.17% -0.67% -2.53% -0.01