chore: Redundant types from docstrings (#206)

* drop redundant types

* update docstrings

Author: SkalskiP

trackers/core/bytetrack/kalman_box_tracker.py

@@ -14,20 +14,20 @@ class ByteTrackKalmanBoxTracker:
     its position.
 
     Attributes:
-        tracker_id (int): Unique identifier for the tracker.
-        number_of_successful_updates (int): Number of times the object has been
+        tracker_id: Unique identifier for the tracker.
+        number_of_successful_updates: Number of times the object has been
             updated successfully.
-        time_since_update (int): Number of frames since the last update.
-        state (np.ndarray): State vector of the bounding box.
-        F (np.ndarray): State transition matrix.
-        H (np.ndarray): Measurement matrix.
-        Q (np.ndarray): Process noise covariance matrix.
-        R (np.ndarray): Measurement noise covariance matrix.
-        P (np.ndarray): Error covariance matrix.
-        count_id (int): Class variable to assign unique IDs to each tracker.
+        time_since_update: Number of frames since the last update.
+        state: State vector of the bounding box.
+        F: State transition matrix.
+        H: Measurement matrix.
+        Q: Process noise covariance matrix.
+        R: Measurement noise covariance matrix.
+        P: Error covariance matrix.
+        count_id: Class variable to assign unique IDs to each tracker.
 
     Args:
-        bbox (np.ndarray): Initial bounding box in the form [x1, y1, x2, y2].
+        bbox: Initial bounding box in the form [x1, y1, x2, y2].
     """
 
     count_id = 0
@@ -38,7 +38,7 @@ def get_next_tracker_id(cls) -> int:
         Class method that returns the next available tracker ID.
 
         Returns:
-            int: The next available tracker ID.
+            The next available tracker ID.
         """
         next_id = cls.count_id
         cls.count_id += 1
@@ -109,7 +109,7 @@ def update(self, bbox: np.ndarray) -> None:
         Updates the state with a new detected bounding box.
 
         Args:
-            bbox (np.ndarray): Detected bounding box in the form [x1, y1, x2, y2].
+            bbox: Detected bounding box in the form [x1, y1, x2, y2].
         """
         self.time_since_update = 0
         self.number_of_successful_updates += 1
@@ -134,7 +134,7 @@ def get_state_bbox(self) -> np.ndarray:
         Returns the current bounding box estimate from the state vector.
 
         Returns:
-            np.ndarray: The bounding box [x1, y1, x2, y2].
+            The bounding box [x1, y1, x2, y2].
         """
         return np.array(
             [

trackers/core/bytetrack/tracker.py

@@ -20,40 +20,28 @@
 
 
 class ByteTrackTracker(BaseTracker):
-    """Implements ByteTrack.
-
-    ByteTrack is a simple, effective, and generic multi-object tracking method
-    that improves upon tracking-by-detection by associating *every* detection box
-    instead of discarding low-score ones. This makes it more robust to occlusions.
-    It uses a two-stage association process and builds on established techniques
-    like the Kalman Filter for motion prediction and the Hungarian algorithm for
-    data association.
+    """Track objects using ByteTrack algorithm with two-stage association.
+    Associates both high and low confidence detections to reduce fragmentation
+    and improve tracking through occlusions.
 
     Args:
-        lost_track_buffer: Number of frames to buffer when a track is lost.
-            Increasing lost_track_buffer enhances occlusion handling, significantly
-            improving tracking through occlusions, but may increase the possibility
-            of ID switching for objects that disappear.
-        frame_rate: Frame rate of the video (frames per second).
-            Used to calculate the maximum time a track can be lost.
-        track_activation_threshold: Detection confidence threshold
-            for track activation. Only detections with confidence above this
-            threshold will create new tracks. Increasing this threshold may
-            reduce false positives but may miss real objects with low confidence.
-        minimum_consecutive_frames: Number of consecutive frames that an object
-            must be tracked before it is considered a 'valid'/'active/ track. Increasing
-            `minimum_consecutive_frames` prevents the creation of accidental tracks
-            from false detection or double detection, but risks missing shorter
-            tracks. Before the tracker is considered valid, it will be assigned
-            `-1` as its `tracker_id`.
-        minimum_iou_threshold: IoU threshold for associating detections to existing tracks.
-            Prevents the association of lower IoU than the threshold between boxes and tracks.
-            A higher value will only associate boxes that have more overlapping area.
-        high_conf_det_threshold: threshold for assigning detections to high probability class.
-            A higher value will classify only higher confidence/probability detections as 'high probability'
-            per the ByteTrack algorithm, which are used in the first similarity step of
-            the algorithm.
-    """  # noqa: E501
+        lost_track_buffer: `int` specifying number of frames to buffer when a
+            track is lost. Increasing this value enhances occlusion handling but
+            may increase ID switching for disappearing objects.
+        frame_rate: `float` specifying video frame rate in frames per second.
+            Used to scale the lost track buffer for consistent tracking across
+            different frame rates.
+        track_activation_threshold: `float` specifying minimum detection
+            confidence to create new tracks. Higher values reduce false
+            positives but may miss low-confidence objects.
+        minimum_consecutive_frames: `int` specifying number of consecutive
+            frames before a track is considered valid. Before reaching this
+            threshold, tracks are assigned `tracker_id` of `-1`.
+        minimum_iou_threshold: `float` specifying IoU threshold for associating
+            detections to existing tracks. Higher values require more overlap.
+        high_conf_det_threshold: `float` specifying threshold for separating
+            high and low confidence detections in the two-stage association.
+    """
 
     def __init__(
         self,
@@ -104,20 +92,19 @@ def update(
         self,
         detections: sv.Detections,
     ) -> sv.Detections:
-        """Updates the tracker state with new detections.
-
-        Performs Kalman Filter prediction, associates detections with existing
-        tracks based on IoU, updates matched tracks, and initializes new
-        tracks for unmatched high-confidence detections.
+        """Update tracker state with new detections and return tracked objects.
+        Performs Kalman filter prediction, two-stage association (high then low
+        confidence), and initializes new tracks for unmatched detections.
 
         Args:
-            detections: The latest set of object detections from a frame.
+            detections: `sv.Detections` containing bounding boxes with shape
+                `(N, 4)` in `(x_min, y_min, x_max, y_max)` format and optional
+                confidence scores.
 
         Returns:
-            A copy of the input detections, augmented with assigned `tracker_id` for
-                each successfully tracked object. Detections not associated with a
-                track will not have a `tracker_id`. The order of the detections is not
-                guaranteed to be the same as the input detections.
+            `sv.Detections` with `tracker_id` assigned for each detection.
+                Unmatched detections have `tracker_id` of `-1`. Detection order
+                may differ from input.
         """
 
         if len(self.tracks) == 0 and len(detections) == 0:
@@ -333,9 +320,8 @@ def _similarity_step(
         return matched_indices, unmatched_tracks, unmatched_detections
 
     def reset(self) -> None:
-        """Resets the tracker's internal state.
-
-        Clears all active tracks and resets the track ID counter.
+        """Reset tracker state by clearing all tracks and resetting ID counter.
+        Call this method when switching to a new video or scene.
         """
         self.tracks = []
         ByteTrackKalmanBoxTracker.count_id = 0

trackers/core/sort/kalman_box_tracker.py

@@ -15,20 +15,20 @@ class SORTKalmanBoxTracker:
     its position.
 
     Attributes:
-        tracker_id (int): Unique identifier for the tracker.
-        number_of_successful_updates (int): Number of times the object has been
+        tracker_id: Unique identifier for the tracker.
+        number_of_successful_updates: Number of times the object has been
             updated successfully.
-        time_since_update (int): Number of frames since the last update.
-        state (np.ndarray): State vector of the bounding box.
-        F (np.ndarray): State transition matrix.
-        H (np.ndarray): Measurement matrix.
-        Q (np.ndarray): Process noise covariance matrix.
-        R (np.ndarray): Measurement noise covariance matrix.
-        P (np.ndarray): Error covariance matrix.
-        count_id (int): Class variable to assign unique IDs to each tracker.
+        time_since_update: Number of frames since the last update.
+        state: State vector of the bounding box.
+        F: State transition matrix.
+        H: Measurement matrix.
+        Q: Process noise covariance matrix.
+        R: Measurement noise covariance matrix.
+        P: Error covariance matrix.
+        count_id: Class variable to assign unique IDs to each tracker.
 
     Args:
-        bbox (np.ndarray): Initial bounding box in the form [x1, y1, x2, y2].
+        bbox: Initial bounding box in the form [x1, y1, x2, y2].
     """
 
     count_id: int = 0
@@ -111,7 +111,7 @@ def update(self, bbox: NDArray[np.float64]) -> None:
         Updates the state with a new detected bounding box.
 
         Args:
-            bbox (np.ndarray): Detected bounding box in the form [x1, y1, x2, y2].
+            bbox: Detected bounding box in the form [x1, y1, x2, y2].
         """
         self.time_since_update = 0
         self.number_of_successful_updates += 1
@@ -140,6 +140,6 @@ def get_state_bbox(self) -> NDArray[np.float32]:
         Returns the current bounding box estimate from the state vector.
 
         Returns:
-            np.ndarray: The bounding box [x1, y1, x2, y2]
+            The bounding box [x1, y1, x2, y2].
         """
         return self.state[:4, 0].flatten().astype(np.float32)

trackers/core/sort/tracker.py

@@ -18,31 +18,25 @@
 
 
 class SORTTracker(BaseTracker):
-    """Implements SORT (Simple Online and Realtime Tracking).
-
-    SORT is a pragmatic approach to multiple object tracking with a focus on
-    simplicity and speed. It uses a Kalman filter for motion prediction and the
-    Hungarian algorithm or simple IOU matching for data association.
+    """Track objects using SORT algorithm with Kalman filter and IoU matching.
+    Provides simple and fast online tracking using only bounding box geometry
+    without appearance features.
 
     Args:
-        lost_track_buffer: Number of frames to buffer when a track is lost.
-            Increasing lost_track_buffer enhances occlusion handling, significantly
-            improving tracking through occlusions, but may increase the possibility
-            of ID switching for objects with similar appearance.
-        frame_rate: Frame rate of the video (frames per second).
-            Used to calculate the maximum time a track can be lost.
-        track_activation_threshold: Detection confidence threshold
-            for track activation. Only detections with confidence above this
-            threshold will create new tracks. Increasing this threshold
-            reduces false positives but may miss real objects with low confidence.
-        minimum_consecutive_frames: Number of consecutive frames that an object
-            must be tracked before it is considered a 'valid' track. Increasing
-            `minimum_consecutive_frames` prevents the creation of accidental tracks
-            from false detection or double detection, but risks missing shorter
-            tracks. Before the tracker is considered valid, it will be assigned
-            `-1` as its `tracker_id`.
-        minimum_iou_threshold: IOU threshold for associating detections to
-            existing tracks.
+        lost_track_buffer: `int` specifying number of frames to buffer when a
+            track is lost. Increasing this value enhances occlusion handling but
+            may increase ID switching for similar objects.
+        frame_rate: `float` specifying video frame rate in frames per second.
+            Used to scale the lost track buffer for consistent tracking across
+            different frame rates.
+        track_activation_threshold: `float` specifying minimum detection
+            confidence to create new tracks. Higher values reduce false
+            positives but may miss low-confidence objects.
+        minimum_consecutive_frames: `int` specifying number of consecutive
+            frames before a track is considered valid. Before reaching this
+            threshold, tracks are assigned `tracker_id` of `-1`.
+        minimum_iou_threshold: `float` specifying IoU threshold for associating
+            detections to existing tracks. Higher values require more overlap.
     """
 
     def __init__(
@@ -120,19 +114,18 @@ def _spawn_new_trackers(
                 self.trackers.append(new_tracker)
 
     def update(self, detections: sv.Detections) -> sv.Detections:
-        """Updates the tracker state with new detections.
-
-        Performs Kalman filter prediction, associates detections with existing
-        trackers based on IOU, updates matched trackers, and initializes new
-        trackers for unmatched high-confidence detections.
+        """Update tracker state with new detections and return tracked objects.
+        Performs Kalman filter prediction, IoU-based association, and initializes
+        new tracks for unmatched high-confidence detections.
 
         Args:
-            detections: The latest set of object detections from a frame.
+            detections: `sv.Detections` containing bounding boxes with shape
+                `(N, 4)` in `(x_min, y_min, x_max, y_max)` format and optional
+                confidence scores.
 
         Returns:
-            A copy of the input detections, augmented with assigned `tracker_id` for
-                each successfully tracked object. Detections not associated with a
-                track will not have a `tracker_id`.
+            `sv.Detections` with `tracker_id` assigned for each detection.
+                Unmatched or immature tracks have `tracker_id` of `-1`.
         """
 
         if len(self.trackers) == 0 and len(detections) == 0:
@@ -180,9 +173,8 @@ def update(self, detections: sv.Detections) -> sv.Detections:
         return updated_detections
 
     def reset(self) -> None:
-        """Resets the tracker's internal state.
-
-        Clears all active tracks and resets the track ID counter.
+        """Reset tracker state by clearing all tracks and resetting ID counter.
+        Call this method when switching to a new video or scene.
         """
         self.trackers = []
         SORTKalmanBoxTracker.count_id = 0

trackers/utils/sort_utils.py

@@ -29,14 +29,14 @@ def get_alive_trackers(
     it was just updated).
 
     Args:
-        trackers (Sequence[KalmanBoxTrackerType]): List of KalmanBoxTracker objects.
-        minimum_consecutive_frames (int): Number of consecutive frames that an object
+        trackers: List of KalmanBoxTracker objects.
+        minimum_consecutive_frames: Number of consecutive frames that an object
             must be tracked before it is considered a 'valid' track.
-        maximum_frames_without_update (int): Maximum number of frames without update
+        maximum_frames_without_update: Maximum number of frames without update
             before a track is considered dead.
 
     Returns:
-        List[KalmanBoxTrackerType]: List of alive trackers.
+        List of alive trackers.
     """
     alive_trackers = []
     for tracker in trackers:
@@ -56,11 +56,12 @@ def get_iou_matrix(
     Build IOU cost matrix between detections and predicted bounding boxes
 
     Args:
-        detection_boxes (np.ndarray): Detected bounding boxes in the
+        trackers: List of KalmanBoxTracker objects.
+        detection_boxes: Detected bounding boxes in the
             form [x1, y1, x2, y2].
 
     Returns:
-        np.ndarray: IOU cost matrix.
+        IOU cost matrix.
     """
     predicted_boxes = np.array([t.get_state_bbox() for t in trackers])
     if len(predicted_boxes) == 0 and len(trackers) > 0:
@@ -88,17 +89,17 @@ def update_detections_with_track_ids(
     it is assigned an ID to the detection that just updated it.
 
     Args:
-        trackers (Sequence[SORTKalmanBoxTracker]): List of SORTKalmanBoxTracker objects.
-        detections (sv.Detections): The latest set of object detections.
-        detection_boxes (np.ndarray): Detected bounding boxes in the
+        trackers: List of SORTKalmanBoxTracker objects.
+        detections: The latest set of object detections.
+        detection_boxes: Detected bounding boxes in the
             form [x1, y1, x2, y2].
-        minimum_iou_threshold (float): IOU threshold for associating detections to
+        minimum_iou_threshold: IOU threshold for associating detections to
             existing tracks.
-        minimum_consecutive_frames (int): Number of consecutive frames that an object
+        minimum_consecutive_frames: Number of consecutive frames that an object
             must be tracked before it is considered a 'valid' track.
 
     Returns:
-        sv.Detections: A copy of the detections with `tracker_id` set
+        A copy of the detections with `tracker_id` set
             for each detection that is tracked.
     """
     # Re-run association in the same way (could also store direct mapping)