Fix models folder issues: bugs and code cleanup

Fixed Issues:
1. tracker.py - Fixed duplicate tracks bug
- Recovered detections were indexed against wrong list
- Recovery function now receives (original_index, detection) pairs
- Returns set of original indices that were recovered
- Prevents creating duplicate tracks for recovered items

2. Code cleanup across all model files:
- tracker.py: Removed unused imports (Path, Union, InferenceError, LogTimer, get_config)
- audio_analyzer.py: Removed unused imports (Dict, ModelLoadError, batch_list)
- visual_analyzer.py: Removed unused import (batch_list) and parameters (prompt, batch_size)
- face_recognizer.py: Removed unused import (validate_image_file)
- body_recognizer.py: Removed unused import (ModelLoadError)
- motion_detector.py: Removed unused imports (Path, Union, ModelLoadError, InferenceError)
- motion_detector.py: Removed dead assignment (_raft_transforms)

Impact:
- Tracker now correctly handles lost track recovery without duplicates
- Cleaner codebase with no unused imports or parameters
- Better code maintainability

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

models/audio_analyzer.py

@@ -14,12 +14,12 @@ Features extracted:
 """
 
 from pathlib import Path
-from typing import List, Optional, Tuple, Dict
+from typing import List, Optional, Tuple
 from dataclasses import dataclass
 import numpy as np
 
 from utils.logger import get_logger, LogTimer
-from utils.helpers import ModelLoadError, InferenceError, normalize_scores, batch_list
+from utils.helpers import InferenceError, normalize_scores
 from config import get_config, ModelConfig
 
 logger = get_logger("models.audio_analyzer")

models/body_recognizer.py

@@ -15,7 +15,7 @@ from dataclasses import dataclass
 import numpy as np
 
 from utils.logger import get_logger, LogTimer
-from utils.helpers import ModelLoadError, InferenceError
+from utils.helpers import InferenceError
 from config import get_config, ModelConfig
 
 logger = get_logger("models.body_recognizer")

models/face_recognizer.py

@@ -14,7 +14,7 @@ from dataclasses import dataclass
 import numpy as np
 
 from utils.logger import get_logger, LogTimer
-from utils.helpers import ModelLoadError, InferenceError, validate_image_file
+from utils.helpers import ModelLoadError, InferenceError
 from config import get_config, ModelConfig
 
 logger = get_logger("models.face_recognizer")

models/motion_detector.py

@@ -10,13 +10,11 @@ Uses RAFT (Recurrent All-Pairs Field Transforms) for high-quality
 optical flow, with fallback to Farneback for speed.
 """
 
-from pathlib import Path
-from typing import List, Optional, Tuple, Union
+from typing import List, Optional, Tuple
 from dataclasses import dataclass
 import numpy as np
 
 from utils.logger import get_logger, LogTimer
-from utils.helpers import ModelLoadError, InferenceError
 from config import get_config, ModelConfig
 
 logger = get_logger("models.motion_detector")
@@ -90,9 +88,6 @@ class MotionDetector:
 
             self.raft_model.eval()
 
-            # Store preprocessing transforms
-            self._raft_transforms = weights.transforms()
-
             logger.info("RAFT model loaded successfully")
 
         except Exception as e:

models/tracker.py

@@ -9,14 +9,11 @@ Multi-object tracking using ByteTrack for:
 ByteTrack uses two-stage association for robust tracking.
 """
 
-from pathlib import Path
-from typing import List, Optional, Dict, Tuple, Union
+from typing import List, Optional, Dict, Tuple
 from dataclasses import dataclass, field
 import numpy as np
 
-from utils.logger import get_logger, LogTimer
-from utils.helpers import InferenceError
-from config import get_config
+from utils.logger import get_logger
 
 logger = get_logger("models.tracker")
 
@@ -162,15 +159,17 @@ class ObjectTracker:
                 self._lost_tracks[track_id] = self._tracks.pop(track_id)
 
         # Try to recover lost tracks with unmatched detections
-        recovered = self._recover_lost_tracks(
-            [(high_conf[i] if i < len(high_conf) else low_conf[i - len(high_conf)])
-             for i in unmatched_dets]
-        )
+        # Build list of (original_index, detection) pairs
+        unmatched_detections = [
+            (i, high_conf[i] if i < len(high_conf) else low_conf[i - len(high_conf)])
+            for i in unmatched_dets
+        ]
+        recovered_indices = self._recover_lost_tracks(unmatched_detections)
 
         # Create new tracks for remaining detections
         new_tracks = []
         for i in unmatched_dets:
-            if i not in recovered:
+            if i not in recovered_indices:
                 det = high_conf[i] if i < len(high_conf) else low_conf[i - len(high_conf)]
                 bbox, conf = det
                 track_id = self._create_track(bbox, conf)
@@ -308,15 +307,23 @@ class ObjectTracker:
 
     def _recover_lost_tracks(
         self,
-        detections: List[Tuple[Tuple[int, int, int, int], float]],
+        detections: List[Tuple[int, Tuple[Tuple[int, int, int, int], float]]],
     ) -> set:
-        """Try to recover lost tracks with unmatched detections."""
+        """
+        Try to recover lost tracks with unmatched detections.
+
+        Args:
+            detections: List of (original_index, (bbox, confidence)) tuples
+
+        Returns:
+            Set of original indices that were successfully recovered
+        """
         recovered = set()
 
         if not self._lost_tracks or not detections:
             return recovered
 
-        for det_idx, (bbox, conf) in enumerate(detections):
+        for orig_idx, (bbox, conf) in detections:
             best_iou = 0
             best_track_id = None
 
@@ -328,7 +335,7 @@ class ObjectTracker:
 
             if best_track_id is not None:
                 self._update_track(best_track_id, bbox, conf)
-                recovered.add(det_idx)
+                recovered.add(orig_idx)  # Add original index, not enumeration index
                 logger.debug(f"Recovered track {best_track_id}")
 
         return recovered

models/visual_analyzer.py

@@ -16,7 +16,7 @@ from dataclasses import dataclass
 import numpy as np
 
 from utils.logger import get_logger, LogTimer
-from utils.helpers import ModelLoadError, InferenceError, batch_list
+from utils.helpers import ModelLoadError, InferenceError
 from config import get_config, ModelConfig
 
 logger = get_logger("models.visual_analyzer")
@@ -180,7 +180,6 @@ Respond with just the emotion."""
     def analyze_frame(
         self,
         image: Union[str, Path, np.ndarray, "PIL.Image.Image"],
-        prompt: Optional[str] = None,
         timestamp: float = 0.0,
     ) -> VisualFeatures:
         """
@@ -188,7 +187,6 @@ Respond with just the emotion."""
 
         Args:
             image: Image path, numpy array, or PIL Image
-            prompt: Custom prompt (uses default if None)
             timestamp: Timestamp for this frame
 
         Returns:
@@ -350,7 +348,6 @@ Respond with just the emotion."""
         self,
         images: List[Union[str, Path, np.ndarray]],
         timestamps: Optional[List[float]] = None,
-        batch_size: int = 4,
     ) -> List[VisualFeatures]:
         """
         Analyze multiple frames in batches.
@@ -358,7 +355,6 @@ Respond with just the emotion."""
         Args:
             images: List of images (paths or arrays)
             timestamps: Timestamps for each image
-            batch_size: Number of images per batch
 
         Returns:
             List of VisualFeatures for each image