IGNF
/

AGarioud nielsr HF Staff commited on
Commit
5e24c27
·
verified ·
1 Parent(s): 71f5f35

Add link to project page and acknowledgement section. (#1)

Browse files

- Add link to project page and acknowledgement section. (cf6e5fbe3c9e23c39d4abbe77334a3f28dd62ec9)


Co-authored-by: Niels Rogge <[email protected]>

Files changed (1) hide show
  1. README.md +39 -89
README.md CHANGED
@@ -1,10 +1,10 @@
1
  ---
2
  license: etalab-2.0
 
3
  tags:
4
  - semantic segmentation
5
  - pytorch
6
  - landcover
7
-
8
  model-index:
9
  - name: FLAIR-HUB_LPIS-I_swinbase-upernet
10
  results:
@@ -14,63 +14,60 @@ model-index:
14
  name: IGNF/FLAIR-HUB/
15
  type: earth-observation-dataset
16
  metrics:
17
- - name: mIoU
18
- type: mIoU
19
  value: 35.76
20
- - name: Overall Accuracy
21
- type: OA
22
  value: 87.189
23
- - name: IoU building
24
- type: IoU
25
  value: 83.86
26
- - name: IoU greenhouse
27
- type: IoU
28
  value: 78.38
29
- - name: IoU swimming pool
30
- type: IoU
31
  value: 61.59
32
- - name: IoU impervious surface
33
- type: IoU
34
  value: 61.59
35
- - name: IoU pervious surface
36
- type: IoU
37
  value: 57.17
38
- - name: IoU bare soil
39
- type: IoU
40
  value: 62.94
41
- - name: IoU water
42
- type: IoU
43
  value: 90.35
44
- - name: IoU snow
45
- type: IoU
46
  value: 63.38
47
- - name: IoU herbaceous vegetation
48
- type: IoU
49
  value: 54.34
50
- - name: IoU agricultural land
51
- type: IoU
52
  value: 57.14
53
- - name: IoU plowed land
54
- type: IoU
55
  value: 34.85
56
- - name: IoU vineyard
57
- type: IoU
58
  value: 33.017
59
- - name: IoU deciduous
60
- type: IoU
61
  value: 71.73
62
- - name: IoU coniferous
63
- type: IoU
64
  value: 62.6
65
- - name: IoU brushwood
66
- type: IoU
67
  value: 30.19
68
-
69
- pipeline_tag: image-segmentation
70
  ---
71
 
72
-
73
-
74
  <div style="font-family:sans-serif; background-color:#F8F5F5; color:black; padding:25px; border-radius:10px; margin:auto; border:0px; ">
75
 
76
  <!-- Collection Section -->
@@ -274,13 +271,13 @@ pipeline_tag: image-segmentation
274
  </div>
275
  </div>
276
 
277
-
278
  ---
279
 
280
  ## General Informations
281
  - **Contact:** [email protected]
282
  - **Code repository:** https://github.com/IGNF/FLAIR-HUB
283
  - **Paper:** https://arxiv.org/abs/2506.07080
 
284
  - **Developed by:** IGN
285
  - **Compute infrastructure:**
286
  - software: python, pytorch-lightning
@@ -289,7 +286,6 @@ pipeline_tag: image-segmentation
289
 
290
  ---
291
 
292
-
293
  ### Training Config Hyperparameters
294
 
295
  ```yaml
@@ -371,53 +367,7 @@ pipeline_tag: image-segmentation
371
  | other arable crops | 22.77 | 37.10 | 32.97 | 42.41 |
372
  | vineyard | 33.02 | 49.64 | 58.03 | 43.37 |
373
  | olive groves | 14.16 | 24.80 | 25.63 | 24.02 |
374
- | fruits orchards | 27.82 | 43.53 | 49.41 | 38.90 |
375
  | nut orchards | 29.83 | 45.95 | 68.55 | 34.56 |
376
  | other permanent crops | 0.27 | 0.53 | 20.92 | 0.27 |
377
- | mixed crops | 5.49 | 10.42 | 25.67 | 6.53 |
378
- | background | 87.62 | 93.40 | 92.01 | 94.84 |
379
-
380
-
381
- ---
382
-
383
-
384
- ## Inference
385
-
386
- <div style="display: flex; justify-content: center; text-align: center; gap: 20px;">
387
-
388
- <div style="flex: 1;">
389
- <p style="margin: 0;">Aerial ROI</p>
390
- <img src="./model_utils/AerialROI.png" alt="AERIAL" style="width: 100%; display: block;" />
391
- </div>
392
- <div style="flex: 1;">
393
- <p style="margin: 0;">Inference ROI</p>
394
- <img src="./model_utils/FLAIR-HUB_LPIS-I_swinbase-upernet_inferenceROI.png" alt="INFERENCE" style="width: 100%; display: block;" />
395
- </div>
396
- </div>
397
-
398
-
399
- ---
400
-
401
-
402
- ## Cite
403
-
404
-
405
- **BibTeX:**
406
-
407
- ```
408
- @article{ign2025flairhub,
409
- doi = {10.48550/arXiv.2506.07080},
410
- url = {https://arxiv.org/abs/2506.07080},
411
- author = {Garioud, Anatol and Giordano, Sébastien and David, Nicolas and Gonthier, Nicolas},
412
- title = {FLAIR-HUB: Large-scale Multimodal Dataset for Land Cover and Crop Mapping},
413
- publisher = {arXiv},
414
- year = {2025}
415
- }
416
- ```
417
-
418
- **APA:**
419
- ```
420
- Anatol Garioud, Sébastien Giordano, Nicolas David, Nicolas Gonthier.
421
- FLAIR-HUB: Large-scale Multimodal Dataset for Land Cover and Crop Mapping. (2025).
422
- DOI: https://doi.org/10.48550/arXiv.2506.07080
423
- ```
 
1
  ---
2
  license: etalab-2.0
3
+ pipeline_tag: image-segmentation
4
  tags:
5
  - semantic segmentation
6
  - pytorch
7
  - landcover
 
8
  model-index:
9
  - name: FLAIR-HUB_LPIS-I_swinbase-upernet
10
  results:
 
14
  name: IGNF/FLAIR-HUB/
15
  type: earth-observation-dataset
16
  metrics:
17
+ - type: mIoU
 
18
  value: 35.76
19
+ name: mIoU
20
+ - type: OA
21
  value: 87.189
22
+ name: Overall Accuracy
23
+ - type: IoU
24
  value: 83.86
25
+ name: IoU building
26
+ - type: IoU
27
  value: 78.38
28
+ name: IoU greenhouse
29
+ - type: IoU
30
  value: 61.59
31
+ name: IoU swimming pool
32
+ - type: IoU
33
  value: 61.59
34
+ name: IoU impervious surface
35
+ - type: IoU
36
  value: 57.17
37
+ name: IoU pervious surface
38
+ - type: IoU
39
  value: 62.94
40
+ name: IoU bare soil
41
+ - type: IoU
42
  value: 90.35
43
+ name: IoU water
44
+ - type: IoU
45
  value: 63.38
46
+ name: IoU snow
47
+ - type: IoU
48
  value: 54.34
49
+ name: IoU herbaceous vegetation
50
+ - type: IoU
51
  value: 57.14
52
+ name: IoU agricultural land
53
+ - type: IoU
54
  value: 34.85
55
+ name: IoU plowed land
56
+ - type: IoU
57
  value: 33.017
58
+ name: IoU vineyard
59
+ - type: IoU
60
  value: 71.73
61
+ name: IoU deciduous
62
+ - type: IoU
63
  value: 62.6
64
+ name: IoU coniferous
65
+ - type: IoU
66
  value: 30.19
67
+ name: IoU brushwood
68
+ library_name: pytorch
69
  ---
70
 
 
 
71
  <div style="font-family:sans-serif; background-color:#F8F5F5; color:black; padding:25px; border-radius:10px; margin:auto; border:0px; ">
72
 
73
  <!-- Collection Section -->
 
271
  </div>
272
  </div>
273
 
 
274
  ---
275
 
276
  ## General Informations
277
  - **Contact:** [email protected]
278
  - **Code repository:** https://github.com/IGNF/FLAIR-HUB
279
  - **Paper:** https://arxiv.org/abs/2506.07080
280
+ - **Project Page:** https://ignf.github.io/FLAIR/FLAIR-HUB/flairhub
281
  - **Developed by:** IGN
282
  - **Compute infrastructure:**
283
  - software: python, pytorch-lightning
 
286
 
287
  ---
288
 
 
289
  ### Training Config Hyperparameters
290
 
291
  ```yaml
 
367
  | other arable crops | 22.77 | 37.10 | 32.97 | 42.41 |
368
  | vineyard | 33.02 | 49.64 | 58.03 | 43.37 |
369
  | olive groves | 14.16 | 24.80 | 25.63 | 24.02 |
370
+ | fruit orchards | 27.82 | 43.53 | 49.41 | 38.90 |
371
  | nut orchards | 29.83 | 45.95 | 68.55 | 34.56 |
372
  | other permanent crops | 0.27 | 0.53 | 20.92 | 0.27 |
373
+ | mixed crops | 5.49 | 10