Downloads · 30 days
34
17% of all-time downloads
MuayThaiLegz/yolo26l-vessel-coastguard-v1
yolo26l-vessel-coastguard-v1 is a object detection model from MuayThaiLegz. Use it when you need objects located in an image. It is set up for ultralytics. The card lists the license as agpl-3.0.
Fine-tuned YOLO26L vessel/maritime detector trained on hand-labeled coast-surveillance footage. Built for the OmniSense Defense ISR prototype.
Downloads · 30 days
34
17% of all-time downloads
All-time downloads
196
Public
Repo size
53.1 MB
Likes
0
Public
Click a slice to open those files.
.pt53.1 MB · 100%
From the Hugging Face model README
Fine-tuned YOLO26L vessel/maritime detector trained on hand-labeled coast-surveillance footage. Built for the OmniSense Defense ISR prototype.
| ID | Name |
|---|---|
| 0 | Cargo Ship |
| 1 | Naval Vessel |
| 2 | Small Boat |
| 3 | Tanker |
| 4 | Unknown |
| 5 | Vessel |
| Metric | All | Small Boat |
|---|---|---|
| mAP50 | 0.875 | 0.910 |
| mAP50-95 | 0.452 | 0.409 |
| Precision | 0.831 | 0.930 |
| Recall | 0.764 | 0.814 |
Trained on 250 frames, validated on 50 frames from DOD_CoastSurveillance.mp4.
from ultralytics import YOLO
model = YOLO("best.pt")
results = model.predict("frame.jpg", imgsz=1280, conf=0.35,
iou=0.45, agnostic_nms=True)
from sahi import AutoDetectionModel
from sahi.predict import get_sliced_prediction
model = AutoDetectionModel.from_pretrained(
model_type="yolov8",
model_path="best.pt",
confidence_threshold=0.30,
device="cuda:0",
)
result = get_sliced_prediction(
"frame.jpg", model,
slice_height=640, slice_width=640,
overlap_height_ratio=0.2, overlap_width_ratio=0.2,
perform_standard_pred=True,
postprocess_type="NMS",
postprocess_match_threshold=0.5,
)
Best_Maritime_Custom.pt (warm-started through run1 → run2 → run4)imgsz=1280; smaller sizes hurt small-vessel recall