OceanGPT-X
Collection
4 items • Updated • 2
Specialized YOLOv5 detection models for fine-grained localization and classification of marine targets. Invoked after the routing stage.
| File | Task | Classes | Architecture |
|---|---|---|---|
sonar_detector/best.pt |
Sonar target detection | 15 sonar categories | YOLOv5 |
fish_detector/best.pt |
Fish species/object detection | Multiple fish classes | YOLOv5 |
coral_detector/best.pt |
Coral species/object detection | Multiple coral classes | YOLOv5 |
Requires YOLOv5.
import torch
# Load detector (e.g., Fish)
model = torch.hub.load("ultralytics/yolov5", "custom", path="fish_detector/best.pt", force_reload=True)
# Inference
results = model("input.jpg", conf=0.5, iou=0.45)
results.print()