MuayThaiLegz commited on
Commit
fd4c0dd
·
verified ·
1 Parent(s): 9fa250e

MilitaryConvoy-YOLO11L — mAP@0.5: 50.0%

Browse files
Files changed (5) hide show
  1. README.md +66 -0
  2. best.pt +3 -0
  3. data.yaml +20 -0
  4. eval_metrics.json +6 -0
  5. last.pt +3 -0
README.md ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ tags:
4
+ - ultralytics
5
+ - yolo
6
+ - yolo11
7
+ - object-detection
8
+ - military-vehicle-detection
9
+ - convoy-detection
10
+ - visdrone
11
+ - aerial-detection
12
+ - OmniSense
13
+ - MediaSense
14
+ - DeSense
15
+ - ConnectiviaLabs
16
+ datasets:
17
+ - visdrone
18
+ - military-vehicle-recognition
19
+ pipeline_tag: object-detection
20
+ library_name: ultralytics
21
+ ---
22
+
23
+ # MilitaryConvoy-YOLO11L — Military Vehicle & Convoy Detection
24
+
25
+ Part of **OmniSense / MediaSense / DeSense** — Connectivia Labs' defense AI platform.
26
+
27
+ ## Performance
28
+ | Metric | Value |
29
+ |--------|-------|
30
+ | **mAP@0.5** | **50.0%** |
31
+ | **mAP@0.5:0.95** | **30.4%** |
32
+ | Precision | 57.0% |
33
+ | Recall | 47.7% |
34
+ | Architecture | YOLO11L |
35
+ | Input size | 640×640 |
36
+ | Training date | 2026-04-05 |
37
+
38
+ ## Classes (15 total)
39
+ **Aerial/civilian (VisDrone):** `pedestrian` · `person` · `bicycle` · `car` · `van` · `truck` · `tricycle` · `awning-tricycle` · `bus` · `motor`
40
+
41
+ **Military (DeSense):** `military-vehicle` · `tank` · `apc` · `afv` · `artillery`
42
+
43
+ ## Training
44
+ - Base: YOLO11L (COCO pretrained)
45
+ - Dataset: VisDrone2019-DET + MilitaryVehicleRecognition v7 (~9,300 images merged)
46
+ - Epochs: 75 | Batch: 32 | Imgsz: 640 | Optimizer: AdamW + cosine LR
47
+ - Augmentation: mosaic, mixup, copy-paste, flips, scale
48
+
49
+ ## Usage
50
+ ```python
51
+ from ultralytics import YOLO
52
+ model = YOLO('MuayThaiLegz/MilitaryConvoy-YOLO11L')
53
+ results = model('aerial_frame.jpg')
54
+
55
+ # Filter military classes only (indices 10-14)
56
+ mil = [b for b in results[0].boxes if int(b.cls) >= 10]
57
+ print(f'Military vehicles: {len(mil)}')
58
+ ```
59
+
60
+ ## DeSense Pipeline
61
+ ```
62
+ Camera/Satellite → MilitaryConvoy-YOLO11L → ByteTrack
63
+ → DBSCAN convoy clustering → Stone Soup fusion → Geofence alert
64
+ ```
65
+
66
+ > **OmniSense** · Pillars: MediaSense · DeSense · Connectivia Labs 🔒
best.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71af003311281eff8d54ef42149ebd2d7d1f3d3e44eca7107dbdf18eab4b5aec
3
+ size 51219993
data.yaml ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ names:
2
+ - pedestrian
3
+ - person
4
+ - bicycle
5
+ - car
6
+ - van
7
+ - truck
8
+ - tricycle
9
+ - awning-tricycle
10
+ - bus
11
+ - motor
12
+ - military-vehicle
13
+ - tank
14
+ - apc
15
+ - afv
16
+ - artillery
17
+ nc: 15
18
+ path: /content/drive/MyDrive/OmniSense/MilitaryConvoy/data/merged
19
+ train: train/images
20
+ val: val/images
eval_metrics.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "mAP50": 0.4995,
3
+ "mAP50_95": 0.3045,
4
+ "precision": 0.5699,
5
+ "recall": 0.4771
6
+ }
last.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:644408cb3cb2221a01992b8096ef04c1d1a85ee11d61cbb26363ede81f705b73
3
+ size 51219993