Add YOLOv8 unified detector with class classification

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

vigilar/detection/yolo.py

@@ -0,0 +1,78 @@
+"""Unified object detection using YOLOv8 via ultralytics."""
+
+import logging
+from pathlib import Path
+
+import numpy as np
+
+from vigilar.detection.person import Detection
+
+log = logging.getLogger(__name__)
+
+# COCO class names for domestic animals
+ANIMAL_CLASSES = {"cat", "dog"}
+
+# COCO class names for wildlife (subset that YOLO can detect)
+WILDLIFE_CLASSES = {"bear", "bird", "horse", "cow", "sheep", "elephant", "zebra", "giraffe"}
+
+# Vehicle class names from COCO
+VEHICLE_CLASSES = {"car", "motorcycle", "bus", "truck", "boat"}
+
+
+class YOLODetector:
+    def __init__(self, model_path: str, confidence_threshold: float = 0.5):
+        self._threshold = confidence_threshold
+        self._model = None
+        self.is_loaded = False
+
+        if Path(model_path).exists():
+            try:
+                from ultralytics import YOLO
+                self._model = YOLO(model_path)
+                self.is_loaded = True
+                log.info("YOLO model loaded from %s", model_path)
+            except Exception as e:
+                log.error("Failed to load YOLO model: %s", e)
+        else:
+            log.warning("YOLO model not found at %s — detection disabled", model_path)
+
+    def detect(self, frame: np.ndarray) -> list[Detection]:
+        if not self.is_loaded or self._model is None:
+            return []
+
+        results = self._model(frame, conf=self._threshold, verbose=False)
+        detections = []
+
+        for result in results:
+            for box in result.boxes:
+                class_id = int(box.cls[0])
+                confidence = float(box.conf[0])
+                class_name = result.names[class_id]
+
+                x1, y1, x2, y2 = box.xyxy[0].tolist()
+                x1, y1, x2, y2 = int(x1), int(y1), int(x2), int(y2)
+                bw, bh = x2 - x1, y2 - y1
+                if bw <= 0 or bh <= 0:
+                    continue
+
+                detections.append(Detection(
+                    class_name=class_name,
+                    class_id=class_id,
+                    confidence=confidence,
+                    bbox=(x1, y1, bw, bh),
+                ))
+
+        return detections
+
+    @staticmethod
+    def classify(detection: Detection) -> str:
+        name = detection.class_name
+        if name == "person":
+            return "person"
+        if name in VEHICLE_CLASSES:
+            return "vehicle"
+        if name in ANIMAL_CLASSES:
+            return "domestic_animal"
+        if name in WILDLIFE_CLASSES:
+            return "wildlife"
+        return "other"