CharlesGaydon commited on
Commit
fbfd857
1 Parent(s): 1cdad4d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +15 -18
README.md CHANGED
@@ -17,28 +17,28 @@ model-index:
17
  metrics:
18
  - name: mIoU
19
  type: mIoU
20
- value: 77.2
21
  - name: IoU Other
22
  type: IoU
23
- value: 52.0
24
  - name: IoU Ground
25
  type: IoU
26
- value: 91.7
27
  - name: IoU Vegetation
28
  type: IoU
29
- value: 93.6
30
  - name: IoU Building
31
  type: IoU
32
- value: 90.6
33
  - name: IoU Water
34
  type: IoU
35
- value: 91.2
36
  - name: IoU Bridge
37
  type: IoU
38
- value: 56.8
39
  - name: IoU Permanent Structure
40
  type: IoU
41
- value: 62.1
42
  - task:
43
  type: semantic-segmentation
44
  dataset:
@@ -206,23 +206,20 @@ The following table gives the class-wise metrics on the test set:
206
 
207
  **Class**|**IoU**|**Accuracy**|**Precision**|**Recall**|**F1**
208
  -----|---|--------|---------|------|---
209
- **other**|52.0|54.9|77.8|54.9|64.4
210
- **ground**|91.7|97.7|93.8|97.7|95.8
211
- **vegetation**|93.6|95.6|98.0|95.6|96.8
212
- **building**|90.6|93.7|96.2|93.7|95.0
213
- **water**|91.2|92.6|97.1|92.6|94.8
214
- **bridge**|56.8|96.1|79.3|78.6|79.0
215
- **permanent structure**|62.1|76.6|78.9|76.6|77.7
216
- -----|---|--------|---------|------|---
217
  **Macro Average**|77.5|86.7|88.7|84.2|86.2
218
 
219
 
220
-
221
  The following illustration gives the resulting confusion matrix :
222
  * Left : normalised acording to rows: rows sum at 100% and the **recall** is on the diagonal of the matrix
223
  * Right : normalised acording to columns: columns sum at 100% and the **precision** is on the diagonal of the matrix
224
 
225
-
226
  <div style="position: relative; text-align: center;">
227
  <p style="margin: 0;">Normalized Confusion Matrices. (a) Recall, (b) Precision)</p>
228
  <img src="FRACTAL-LidarHD_7cl_randlanet-recall_confusion_matrix.excalidraw.png" alt="Confusion matrices" style="width: 70%; display: block; margin: 0 auto;"/>
 
17
  metrics:
18
  - name: mIoU
19
  type: mIoU
20
+ value: 77.5
21
  - name: IoU Other
22
  type: IoU
23
+ value: 47.5
24
  - name: IoU Ground
25
  type: IoU
26
+ value: 91.9
27
  - name: IoU Vegetation
28
  type: IoU
29
+ value: 93.8
30
  - name: IoU Building
31
  type: IoU
32
+ value: 90.4
33
  - name: IoU Water
34
  type: IoU
35
+ value: 90.1
36
  - name: IoU Bridge
37
  type: IoU
38
+ value: 65.2
39
  - name: IoU Permanent Structure
40
  type: IoU
41
+ value: 63.5
42
  - task:
43
  type: semantic-segmentation
44
  dataset:
 
206
 
207
  **Class**|**IoU**|**Accuracy**|**Precision**|**Recall**|**F1**
208
  -----|---|--------|---------|------|---
209
+ **other**|47.5|54.9|77.8|54.9|64.4
210
+ **ground**|91.9|97.7|93.8|97.7|95.8
211
+ **vegetation**|93.8|95.6|98.0|95.6|96.8
212
+ **building**|90.4|93.7|96.2|93.7|95.0
213
+ **water**|90.1|92.6|97.1|92.6|94.8
214
+ **bridge**|65.2|96.1|79.3|78.6|79.0
215
+ **permanent structure**|63.5|76.6|78.9|76.6|77.7
 
216
  **Macro Average**|77.5|86.7|88.7|84.2|86.2
217
 
218
 
 
219
  The following illustration gives the resulting confusion matrix :
220
  * Left : normalised acording to rows: rows sum at 100% and the **recall** is on the diagonal of the matrix
221
  * Right : normalised acording to columns: columns sum at 100% and the **precision** is on the diagonal of the matrix
222
 
 
223
  <div style="position: relative; text-align: center;">
224
  <p style="margin: 0;">Normalized Confusion Matrices. (a) Recall, (b) Precision)</p>
225
  <img src="FRACTAL-LidarHD_7cl_randlanet-recall_confusion_matrix.excalidraw.png" alt="Confusion matrices" style="width: 70%; display: block; margin: 0 auto;"/>