Skip to content

Référence pour ultralytics/trackers/utils/kalman_filter.py

Note

Ce fichier est disponible à l'adresse https://github.com/ultralytics/ ultralytics/blob/main/ ultralytics/trackers/utils/kalman_filter .py. Si tu repères un problème, aide à le corriger en contribuant à une Pull Request 🛠️. Merci 🙏 !



ultralytics.trackers.utils.kalman_filter.KalmanFilterXYAH

Pour bytetrack. Un filtre de Kalman simple pour le suivi des boîtes de délimitation dans l'espace de l'image.

L'espace d'état à 8 dimensions (x, y, a, h, vx, vy, va, vh) contient la position centrale de la boîte englobante (x, y), le rapport d'aspect a, la hauteur h et leurs vitesses respectives. a, la hauteur h et leurs vitesses respectives.

Le mouvement de l'objet suit un modèle de vitesse constante. L'emplacement de la boîte englobante (x, y, a, h) est considéré comme une observation directe de l'espace d'état (modèle d'observation linéaire). l'espace d'état (modèle d'observation linéaire).

Code source dans ultralytics/trackers/utils/kalman_filter.py
class KalmanFilterXYAH:
    """
    For bytetrack. A simple Kalman filter for tracking bounding boxes in image space.

    The 8-dimensional state space (x, y, a, h, vx, vy, va, vh) contains the bounding box center position (x, y), aspect
    ratio a, height h, and their respective velocities.

    Object motion follows a constant velocity model. The bounding box location (x, y, a, h) is taken as direct
    observation of the state space (linear observation model).
    """

    def __init__(self):
        """Initialize Kalman filter model matrices with motion and observation uncertainty weights."""
        ndim, dt = 4, 1.0

        # Create Kalman filter model matrices
        self._motion_mat = np.eye(2 * ndim, 2 * ndim)
        for i in range(ndim):
            self._motion_mat[i, ndim + i] = dt
        self._update_mat = np.eye(ndim, 2 * ndim)

        # Motion and observation uncertainty are chosen relative to the current state estimate. These weights control
        # the amount of uncertainty in the model.
        self._std_weight_position = 1.0 / 20
        self._std_weight_velocity = 1.0 / 160

    def initiate(self, measurement: np.ndarray) -> tuple:
        """
        Create track from unassociated measurement.

        Args:
            measurement (ndarray): Bounding box coordinates (x, y, a, h) with center position (x, y), aspect ratio a,
                and height h.

        Returns:
            (tuple[ndarray, ndarray]): Returns the mean vector (8 dimensional) and covariance matrix (8x8 dimensional)
                of the new track. Unobserved velocities are initialized to 0 mean.
        """
        mean_pos = measurement
        mean_vel = np.zeros_like(mean_pos)
        mean = np.r_[mean_pos, mean_vel]

        std = [
            2 * self._std_weight_position * measurement[3],
            2 * self._std_weight_position * measurement[3],
            1e-2,
            2 * self._std_weight_position * measurement[3],
            10 * self._std_weight_velocity * measurement[3],
            10 * self._std_weight_velocity * measurement[3],
            1e-5,
            10 * self._std_weight_velocity * measurement[3],
        ]
        covariance = np.diag(np.square(std))
        return mean, covariance

    def predict(self, mean: np.ndarray, covariance: np.ndarray) -> tuple:
        """
        Run Kalman filter prediction step.

        Args:
            mean (ndarray): The 8 dimensional mean vector of the object state at the previous time step.
            covariance (ndarray): The 8x8 dimensional covariance matrix of the object state at the previous time step.

        Returns:
            (tuple[ndarray, ndarray]): Returns the mean vector and covariance matrix of the predicted state. Unobserved
                velocities are initialized to 0 mean.
        """
        std_pos = [
            self._std_weight_position * mean[3],
            self._std_weight_position * mean[3],
            1e-2,
            self._std_weight_position * mean[3],
        ]
        std_vel = [
            self._std_weight_velocity * mean[3],
            self._std_weight_velocity * mean[3],
            1e-5,
            self._std_weight_velocity * mean[3],
        ]
        motion_cov = np.diag(np.square(np.r_[std_pos, std_vel]))

        mean = np.dot(mean, self._motion_mat.T)
        covariance = np.linalg.multi_dot((self._motion_mat, covariance, self._motion_mat.T)) + motion_cov

        return mean, covariance

    def project(self, mean: np.ndarray, covariance: np.ndarray) -> tuple:
        """
        Project state distribution to measurement space.

        Args:
            mean (ndarray): The state's mean vector (8 dimensional array).
            covariance (ndarray): The state's covariance matrix (8x8 dimensional).

        Returns:
            (tuple[ndarray, ndarray]): Returns the projected mean and covariance matrix of the given state estimate.
        """
        std = [
            self._std_weight_position * mean[3],
            self._std_weight_position * mean[3],
            1e-1,
            self._std_weight_position * mean[3],
        ]
        innovation_cov = np.diag(np.square(std))

        mean = np.dot(self._update_mat, mean)
        covariance = np.linalg.multi_dot((self._update_mat, covariance, self._update_mat.T))
        return mean, covariance + innovation_cov

    def multi_predict(self, mean: np.ndarray, covariance: np.ndarray) -> tuple:
        """
        Run Kalman filter prediction step (Vectorized version).

        Args:
            mean (ndarray): The Nx8 dimensional mean matrix of the object states at the previous time step.
            covariance (ndarray): The Nx8x8 covariance matrix of the object states at the previous time step.

        Returns:
            (tuple[ndarray, ndarray]): Returns the mean vector and covariance matrix of the predicted state. Unobserved
                velocities are initialized to 0 mean.
        """
        std_pos = [
            self._std_weight_position * mean[:, 3],
            self._std_weight_position * mean[:, 3],
            1e-2 * np.ones_like(mean[:, 3]),
            self._std_weight_position * mean[:, 3],
        ]
        std_vel = [
            self._std_weight_velocity * mean[:, 3],
            self._std_weight_velocity * mean[:, 3],
            1e-5 * np.ones_like(mean[:, 3]),
            self._std_weight_velocity * mean[:, 3],
        ]
        sqr = np.square(np.r_[std_pos, std_vel]).T

        motion_cov = [np.diag(sqr[i]) for i in range(len(mean))]
        motion_cov = np.asarray(motion_cov)

        mean = np.dot(mean, self._motion_mat.T)
        left = np.dot(self._motion_mat, covariance).transpose((1, 0, 2))
        covariance = np.dot(left, self._motion_mat.T) + motion_cov

        return mean, covariance

    def update(self, mean: np.ndarray, covariance: np.ndarray, measurement: np.ndarray) -> tuple:
        """
        Run Kalman filter correction step.

        Args:
            mean (ndarray): The predicted state's mean vector (8 dimensional).
            covariance (ndarray): The state's covariance matrix (8x8 dimensional).
            measurement (ndarray): The 4 dimensional measurement vector (x, y, a, h), where (x, y) is the center
                position, a the aspect ratio, and h the height of the bounding box.

        Returns:
            (tuple[ndarray, ndarray]): Returns the measurement-corrected state distribution.
        """
        projected_mean, projected_cov = self.project(mean, covariance)

        chol_factor, lower = scipy.linalg.cho_factor(projected_cov, lower=True, check_finite=False)
        kalman_gain = scipy.linalg.cho_solve(
            (chol_factor, lower), np.dot(covariance, self._update_mat.T).T, check_finite=False
        ).T
        innovation = measurement - projected_mean

        new_mean = mean + np.dot(innovation, kalman_gain.T)
        new_covariance = covariance - np.linalg.multi_dot((kalman_gain, projected_cov, kalman_gain.T))
        return new_mean, new_covariance

    def gating_distance(
        self,
        mean: np.ndarray,
        covariance: np.ndarray,
        measurements: np.ndarray,
        only_position: bool = False,
        metric: str = "maha",
    ) -> np.ndarray:
        """
        Compute gating distance between state distribution and measurements. A suitable distance threshold can be
        obtained from `chi2inv95`. If `only_position` is False, the chi-square distribution has 4 degrees of freedom,
        otherwise 2.

        Args:
            mean (ndarray): Mean vector over the state distribution (8 dimensional).
            covariance (ndarray): Covariance of the state distribution (8x8 dimensional).
            measurements (ndarray): An Nx4 matrix of N measurements, each in format (x, y, a, h) where (x, y)
                is the bounding box center position, a the aspect ratio, and h the height.
            only_position (bool, optional): If True, distance computation is done with respect to the bounding box
                center position only. Defaults to False.
            metric (str, optional): The metric to use for calculating the distance. Options are 'gaussian' for the
                squared Euclidean distance and 'maha' for the squared Mahalanobis distance. Defaults to 'maha'.

        Returns:
            (np.ndarray): Returns an array of length N, where the i-th element contains the squared distance between
                (mean, covariance) and `measurements[i]`.
        """
        mean, covariance = self.project(mean, covariance)
        if only_position:
            mean, covariance = mean[:2], covariance[:2, :2]
            measurements = measurements[:, :2]

        d = measurements - mean
        if metric == "gaussian":
            return np.sum(d * d, axis=1)
        elif metric == "maha":
            cholesky_factor = np.linalg.cholesky(covariance)
            z = scipy.linalg.solve_triangular(cholesky_factor, d.T, lower=True, check_finite=False, overwrite_b=True)
            return np.sum(z * z, axis=0)  # square maha
        else:
            raise ValueError("Invalid distance metric")

__init__()

Initialise les matrices de modèle du filtre de Kalman avec les poids d'incertitude du mouvement et de l'observation.

Code source dans ultralytics/trackers/utils/kalman_filter.py
def __init__(self):
    """Initialize Kalman filter model matrices with motion and observation uncertainty weights."""
    ndim, dt = 4, 1.0

    # Create Kalman filter model matrices
    self._motion_mat = np.eye(2 * ndim, 2 * ndim)
    for i in range(ndim):
        self._motion_mat[i, ndim + i] = dt
    self._update_mat = np.eye(ndim, 2 * ndim)

    # Motion and observation uncertainty are chosen relative to the current state estimate. These weights control
    # the amount of uncertainty in the model.
    self._std_weight_position = 1.0 / 20
    self._std_weight_velocity = 1.0 / 160

gating_distance(mean, covariance, measurements, only_position=False, metric='maha')

Calcule la distance entre la distribution des états et les mesures. Un seuil de distance approprié peut être obtenu à partir chi2inv95. Si only_position est Faux, la distribution du khi-deux a 4 degrés de liberté, sinon 2.

Paramètres :

Nom Type Description DĂ©faut
mean ndarray

Vecteur moyen de la distribution des Ă©tats (8 dimensions).

requis
covariance ndarray

Covariance de la distribution des Ă©tats (8x8 dimensions).

requis
measurements ndarray

Une matrice Nx4 de N mesures, chacune au format (x, y, a, h) où (x, y) est la position centrale de la boîte englobante, a le rapport d'aspect et h la hauteur.

requis
only_position bool

Si True, le calcul de la distance se fait par rapport à la position centrale de la boîte de délimitation. uniquement par rapport à la position centrale de la boîte englobante. La valeur par défaut est False.

False
metric str

La métrique à utiliser pour calculer la distance. Les options sont 'gaussien' pour la distance euclidienne au carré et 'maha' pour la distance de Mahalanobis au carré. pour la distance euclidienne au carré et 'maha' pour la distance de Mahalanobis au carré. La valeur par défaut est 'maha'.

'maha'

Retourne :

Type Description
ndarray

Renvoie un tableau de longueur N, où le i-ème élément contient la distance au carré entre (moyenne, covariance) et measurements[i].

Code source dans ultralytics/trackers/utils/kalman_filter.py
def gating_distance(
    self,
    mean: np.ndarray,
    covariance: np.ndarray,
    measurements: np.ndarray,
    only_position: bool = False,
    metric: str = "maha",
) -> np.ndarray:
    """
    Compute gating distance between state distribution and measurements. A suitable distance threshold can be
    obtained from `chi2inv95`. If `only_position` is False, the chi-square distribution has 4 degrees of freedom,
    otherwise 2.

    Args:
        mean (ndarray): Mean vector over the state distribution (8 dimensional).
        covariance (ndarray): Covariance of the state distribution (8x8 dimensional).
        measurements (ndarray): An Nx4 matrix of N measurements, each in format (x, y, a, h) where (x, y)
            is the bounding box center position, a the aspect ratio, and h the height.
        only_position (bool, optional): If True, distance computation is done with respect to the bounding box
            center position only. Defaults to False.
        metric (str, optional): The metric to use for calculating the distance. Options are 'gaussian' for the
            squared Euclidean distance and 'maha' for the squared Mahalanobis distance. Defaults to 'maha'.

    Returns:
        (np.ndarray): Returns an array of length N, where the i-th element contains the squared distance between
            (mean, covariance) and `measurements[i]`.
    """
    mean, covariance = self.project(mean, covariance)
    if only_position:
        mean, covariance = mean[:2], covariance[:2, :2]
        measurements = measurements[:, :2]

    d = measurements - mean
    if metric == "gaussian":
        return np.sum(d * d, axis=1)
    elif metric == "maha":
        cholesky_factor = np.linalg.cholesky(covariance)
        z = scipy.linalg.solve_triangular(cholesky_factor, d.T, lower=True, check_finite=False, overwrite_b=True)
        return np.sum(z * z, axis=0)  # square maha
    else:
        raise ValueError("Invalid distance metric")

initiate(measurement)

Crée une piste à partir de mesures non associées.

Paramètres :

Nom Type Description DĂ©faut
measurement ndarray

Coordonnées de la boîte de délimitation (x, y, a, h) avec la position centrale (x, y), le rapport d'aspect a, et la hauteur h.

requis

Retourne :

Type Description
tuple[ndarray, ndarray]

Renvoie le vecteur moyen (8 dimensions) et la matrice de covariance (8x8 dimensions) de la nouvelle piste. Les vitesses non observées sont initialisées à 0 moyenne.

Code source dans ultralytics/trackers/utils/kalman_filter.py
def initiate(self, measurement: np.ndarray) -> tuple:
    """
    Create track from unassociated measurement.

    Args:
        measurement (ndarray): Bounding box coordinates (x, y, a, h) with center position (x, y), aspect ratio a,
            and height h.

    Returns:
        (tuple[ndarray, ndarray]): Returns the mean vector (8 dimensional) and covariance matrix (8x8 dimensional)
            of the new track. Unobserved velocities are initialized to 0 mean.
    """
    mean_pos = measurement
    mean_vel = np.zeros_like(mean_pos)
    mean = np.r_[mean_pos, mean_vel]

    std = [
        2 * self._std_weight_position * measurement[3],
        2 * self._std_weight_position * measurement[3],
        1e-2,
        2 * self._std_weight_position * measurement[3],
        10 * self._std_weight_velocity * measurement[3],
        10 * self._std_weight_velocity * measurement[3],
        1e-5,
        10 * self._std_weight_velocity * measurement[3],
    ]
    covariance = np.diag(np.square(std))
    return mean, covariance

multi_predict(mean, covariance)

Exécute l'étape de prédiction du filtre de Kalman (version vectorisée).

Paramètres :

Nom Type Description DĂ©faut
mean ndarray

La matrice moyenne de dimension Nx8 des états de l'objet au pas de temps précédent.

requis
covariance ndarray

La matrice de covariance Nx8x8 des états de l'objet au pas de temps précédent.

requis

Retourne :

Type Description
tuple[ndarray, ndarray]

Renvoie le vecteur moyen et la matrice de covariance de l'état prédit. Les vitesses sont initialisées à une moyenne de 0.

Code source dans ultralytics/trackers/utils/kalman_filter.py
def multi_predict(self, mean: np.ndarray, covariance: np.ndarray) -> tuple:
    """
    Run Kalman filter prediction step (Vectorized version).

    Args:
        mean (ndarray): The Nx8 dimensional mean matrix of the object states at the previous time step.
        covariance (ndarray): The Nx8x8 covariance matrix of the object states at the previous time step.

    Returns:
        (tuple[ndarray, ndarray]): Returns the mean vector and covariance matrix of the predicted state. Unobserved
            velocities are initialized to 0 mean.
    """
    std_pos = [
        self._std_weight_position * mean[:, 3],
        self._std_weight_position * mean[:, 3],
        1e-2 * np.ones_like(mean[:, 3]),
        self._std_weight_position * mean[:, 3],
    ]
    std_vel = [
        self._std_weight_velocity * mean[:, 3],
        self._std_weight_velocity * mean[:, 3],
        1e-5 * np.ones_like(mean[:, 3]),
        self._std_weight_velocity * mean[:, 3],
    ]
    sqr = np.square(np.r_[std_pos, std_vel]).T

    motion_cov = [np.diag(sqr[i]) for i in range(len(mean))]
    motion_cov = np.asarray(motion_cov)

    mean = np.dot(mean, self._motion_mat.T)
    left = np.dot(self._motion_mat, covariance).transpose((1, 0, 2))
    covariance = np.dot(left, self._motion_mat.T) + motion_cov

    return mean, covariance

predict(mean, covariance)

Exécute l'étape de prédiction du filtre de Kalman.

Paramètres :

Nom Type Description DĂ©faut
mean ndarray

Le vecteur moyen à 8 dimensions de l'état de l'objet au pas de temps précédent.

requis
covariance ndarray

La matrice de covariance de dimension 8x8 de l'état de l'objet au pas de temps précédent.

requis

Retourne :

Type Description
tuple[ndarray, ndarray]

Renvoie le vecteur moyen et la matrice de covariance de l'état prédit. Les vitesses sont initialisées à une moyenne de 0.

Code source dans ultralytics/trackers/utils/kalman_filter.py
def predict(self, mean: np.ndarray, covariance: np.ndarray) -> tuple:
    """
    Run Kalman filter prediction step.

    Args:
        mean (ndarray): The 8 dimensional mean vector of the object state at the previous time step.
        covariance (ndarray): The 8x8 dimensional covariance matrix of the object state at the previous time step.

    Returns:
        (tuple[ndarray, ndarray]): Returns the mean vector and covariance matrix of the predicted state. Unobserved
            velocities are initialized to 0 mean.
    """
    std_pos = [
        self._std_weight_position * mean[3],
        self._std_weight_position * mean[3],
        1e-2,
        self._std_weight_position * mean[3],
    ]
    std_vel = [
        self._std_weight_velocity * mean[3],
        self._std_weight_velocity * mean[3],
        1e-5,
        self._std_weight_velocity * mean[3],
    ]
    motion_cov = np.diag(np.square(np.r_[std_pos, std_vel]))

    mean = np.dot(mean, self._motion_mat.T)
    covariance = np.linalg.multi_dot((self._motion_mat, covariance, self._motion_mat.T)) + motion_cov

    return mean, covariance

project(mean, covariance)

Projette la distribution des Ă©tats dans l'espace de mesure.

Paramètres :

Nom Type Description DĂ©faut
mean ndarray

Le vecteur moyen de l'Ă©tat (tableau Ă  8 dimensions).

requis
covariance ndarray

La matrice de covariance de l'Ă©tat (8x8 dimensions).

requis

Retourne :

Type Description
tuple[ndarray, ndarray]

Renvoie la moyenne et la matrice de covariance projetées de l'estimation d'état donnée.

Code source dans ultralytics/trackers/utils/kalman_filter.py
def project(self, mean: np.ndarray, covariance: np.ndarray) -> tuple:
    """
    Project state distribution to measurement space.

    Args:
        mean (ndarray): The state's mean vector (8 dimensional array).
        covariance (ndarray): The state's covariance matrix (8x8 dimensional).

    Returns:
        (tuple[ndarray, ndarray]): Returns the projected mean and covariance matrix of the given state estimate.
    """
    std = [
        self._std_weight_position * mean[3],
        self._std_weight_position * mean[3],
        1e-1,
        self._std_weight_position * mean[3],
    ]
    innovation_cov = np.diag(np.square(std))

    mean = np.dot(self._update_mat, mean)
    covariance = np.linalg.multi_dot((self._update_mat, covariance, self._update_mat.T))
    return mean, covariance + innovation_cov

update(mean, covariance, measurement)

Exécute l'étape de correction du filtre de Kalman.

Paramètres :

Nom Type Description DĂ©faut
mean ndarray

Le vecteur moyen de l'état prédit (8 dimensions).

requis
covariance ndarray

La matrice de covariance de l'Ă©tat (8x8 dimensions).

requis
measurement ndarray

Le vecteur de mesure quadridimensionnel (x, y, a, h), où (x, y) est la position du centre, a le rapport d'aspect et h la hauteur de la boîte de délimitation. la position centrale, a le rapport d'aspect et h la hauteur de la boîte de délimitation.

requis

Retourne :

Type Description
tuple[ndarray, ndarray]

Renvoie la distribution d'état corrigée de la mesure.

Code source dans ultralytics/trackers/utils/kalman_filter.py
def update(self, mean: np.ndarray, covariance: np.ndarray, measurement: np.ndarray) -> tuple:
    """
    Run Kalman filter correction step.

    Args:
        mean (ndarray): The predicted state's mean vector (8 dimensional).
        covariance (ndarray): The state's covariance matrix (8x8 dimensional).
        measurement (ndarray): The 4 dimensional measurement vector (x, y, a, h), where (x, y) is the center
            position, a the aspect ratio, and h the height of the bounding box.

    Returns:
        (tuple[ndarray, ndarray]): Returns the measurement-corrected state distribution.
    """
    projected_mean, projected_cov = self.project(mean, covariance)

    chol_factor, lower = scipy.linalg.cho_factor(projected_cov, lower=True, check_finite=False)
    kalman_gain = scipy.linalg.cho_solve(
        (chol_factor, lower), np.dot(covariance, self._update_mat.T).T, check_finite=False
    ).T
    innovation = measurement - projected_mean

    new_mean = mean + np.dot(innovation, kalman_gain.T)
    new_covariance = covariance - np.linalg.multi_dot((kalman_gain, projected_cov, kalman_gain.T))
    return new_mean, new_covariance



ultralytics.trackers.utils.kalman_filter.KalmanFilterXYWH

Bases : KalmanFilterXYAH

Pour BoT-SORT. Un filtre de Kalman simple pour le suivi des boîtes de délimitation dans l'espace de l'image.

L'espace d'état à 8 dimensions (x, y, w, h, vx, vy, vw, vh) contient la position centrale de la boîte englobante (x, y), sa largeur, sa hauteur et leurs vitesses respectives. w, la hauteur h, et leurs vitesses respectives.

Le mouvement de l'objet suit un modèle de vitesse constante. L'emplacement de la boîte englobante (x, y, w, h) est considéré comme une observation directe de l'espace d'état (modèle d'observation linéaire). l'espace d'état (modèle d'observation linéaire).

Code source dans ultralytics/trackers/utils/kalman_filter.py
class KalmanFilterXYWH(KalmanFilterXYAH):
    """
    For BoT-SORT. A simple Kalman filter for tracking bounding boxes in image space.

    The 8-dimensional state space (x, y, w, h, vx, vy, vw, vh) contains the bounding box center position (x, y), width
    w, height h, and their respective velocities.

    Object motion follows a constant velocity model. The bounding box location (x, y, w, h) is taken as direct
    observation of the state space (linear observation model).
    """

    def initiate(self, measurement: np.ndarray) -> tuple:
        """
        Create track from unassociated measurement.

        Args:
            measurement (ndarray): Bounding box coordinates (x, y, w, h) with center position (x, y), width, and height.

        Returns:
            (tuple[ndarray, ndarray]): Returns the mean vector (8 dimensional) and covariance matrix (8x8 dimensional)
                of the new track. Unobserved velocities are initialized to 0 mean.
        """
        mean_pos = measurement
        mean_vel = np.zeros_like(mean_pos)
        mean = np.r_[mean_pos, mean_vel]

        std = [
            2 * self._std_weight_position * measurement[2],
            2 * self._std_weight_position * measurement[3],
            2 * self._std_weight_position * measurement[2],
            2 * self._std_weight_position * measurement[3],
            10 * self._std_weight_velocity * measurement[2],
            10 * self._std_weight_velocity * measurement[3],
            10 * self._std_weight_velocity * measurement[2],
            10 * self._std_weight_velocity * measurement[3],
        ]
        covariance = np.diag(np.square(std))
        return mean, covariance

    def predict(self, mean, covariance) -> tuple:
        """
        Run Kalman filter prediction step.

        Args:
            mean (ndarray): The 8 dimensional mean vector of the object state at the previous time step.
            covariance (ndarray): The 8x8 dimensional covariance matrix of the object state at the previous time step.

        Returns:
            (tuple[ndarray, ndarray]): Returns the mean vector and covariance matrix of the predicted state. Unobserved
                velocities are initialized to 0 mean.
        """
        std_pos = [
            self._std_weight_position * mean[2],
            self._std_weight_position * mean[3],
            self._std_weight_position * mean[2],
            self._std_weight_position * mean[3],
        ]
        std_vel = [
            self._std_weight_velocity * mean[2],
            self._std_weight_velocity * mean[3],
            self._std_weight_velocity * mean[2],
            self._std_weight_velocity * mean[3],
        ]
        motion_cov = np.diag(np.square(np.r_[std_pos, std_vel]))

        mean = np.dot(mean, self._motion_mat.T)
        covariance = np.linalg.multi_dot((self._motion_mat, covariance, self._motion_mat.T)) + motion_cov

        return mean, covariance

    def project(self, mean, covariance) -> tuple:
        """
        Project state distribution to measurement space.

        Args:
            mean (ndarray): The state's mean vector (8 dimensional array).
            covariance (ndarray): The state's covariance matrix (8x8 dimensional).

        Returns:
            (tuple[ndarray, ndarray]): Returns the projected mean and covariance matrix of the given state estimate.
        """
        std = [
            self._std_weight_position * mean[2],
            self._std_weight_position * mean[3],
            self._std_weight_position * mean[2],
            self._std_weight_position * mean[3],
        ]
        innovation_cov = np.diag(np.square(std))

        mean = np.dot(self._update_mat, mean)
        covariance = np.linalg.multi_dot((self._update_mat, covariance, self._update_mat.T))
        return mean, covariance + innovation_cov

    def multi_predict(self, mean, covariance) -> tuple:
        """
        Run Kalman filter prediction step (Vectorized version).

        Args:
            mean (ndarray): The Nx8 dimensional mean matrix of the object states at the previous time step.
            covariance (ndarray): The Nx8x8 covariance matrix of the object states at the previous time step.

        Returns:
            (tuple[ndarray, ndarray]): Returns the mean vector and covariance matrix of the predicted state. Unobserved
                velocities are initialized to 0 mean.
        """
        std_pos = [
            self._std_weight_position * mean[:, 2],
            self._std_weight_position * mean[:, 3],
            self._std_weight_position * mean[:, 2],
            self._std_weight_position * mean[:, 3],
        ]
        std_vel = [
            self._std_weight_velocity * mean[:, 2],
            self._std_weight_velocity * mean[:, 3],
            self._std_weight_velocity * mean[:, 2],
            self._std_weight_velocity * mean[:, 3],
        ]
        sqr = np.square(np.r_[std_pos, std_vel]).T

        motion_cov = [np.diag(sqr[i]) for i in range(len(mean))]
        motion_cov = np.asarray(motion_cov)

        mean = np.dot(mean, self._motion_mat.T)
        left = np.dot(self._motion_mat, covariance).transpose((1, 0, 2))
        covariance = np.dot(left, self._motion_mat.T) + motion_cov

        return mean, covariance

    def update(self, mean, covariance, measurement) -> tuple:
        """
        Run Kalman filter correction step.

        Args:
            mean (ndarray): The predicted state's mean vector (8 dimensional).
            covariance (ndarray): The state's covariance matrix (8x8 dimensional).
            measurement (ndarray): The 4 dimensional measurement vector (x, y, w, h), where (x, y) is the center
                position, w the width, and h the height of the bounding box.

        Returns:
            (tuple[ndarray, ndarray]): Returns the measurement-corrected state distribution.
        """
        return super().update(mean, covariance, measurement)

initiate(measurement)

Crée une piste à partir de mesures non associées.

Paramètres :

Nom Type Description DĂ©faut
measurement ndarray

Coordonnées du cadre (x, y, w, h) avec la position centrale (x, y), la largeur et la hauteur.

requis

Retourne :

Type Description
tuple[ndarray, ndarray]

Renvoie le vecteur moyen (8 dimensions) et la matrice de covariance (8x8 dimensions) de la nouvelle piste. Les vitesses non observées sont initialisées à 0 moyenne.

Code source dans ultralytics/trackers/utils/kalman_filter.py
def initiate(self, measurement: np.ndarray) -> tuple:
    """
    Create track from unassociated measurement.

    Args:
        measurement (ndarray): Bounding box coordinates (x, y, w, h) with center position (x, y), width, and height.

    Returns:
        (tuple[ndarray, ndarray]): Returns the mean vector (8 dimensional) and covariance matrix (8x8 dimensional)
            of the new track. Unobserved velocities are initialized to 0 mean.
    """
    mean_pos = measurement
    mean_vel = np.zeros_like(mean_pos)
    mean = np.r_[mean_pos, mean_vel]

    std = [
        2 * self._std_weight_position * measurement[2],
        2 * self._std_weight_position * measurement[3],
        2 * self._std_weight_position * measurement[2],
        2 * self._std_weight_position * measurement[3],
        10 * self._std_weight_velocity * measurement[2],
        10 * self._std_weight_velocity * measurement[3],
        10 * self._std_weight_velocity * measurement[2],
        10 * self._std_weight_velocity * measurement[3],
    ]
    covariance = np.diag(np.square(std))
    return mean, covariance

multi_predict(mean, covariance)

Exécute l'étape de prédiction du filtre de Kalman (version vectorisée).

Paramètres :

Nom Type Description DĂ©faut
mean ndarray

La matrice moyenne de dimension Nx8 des états de l'objet au pas de temps précédent.

requis
covariance ndarray

La matrice de covariance Nx8x8 des états de l'objet au pas de temps précédent.

requis

Retourne :

Type Description
tuple[ndarray, ndarray]

Renvoie le vecteur moyen et la matrice de covariance de l'état prédit. Les vitesses sont initialisées à une moyenne de 0.

Code source dans ultralytics/trackers/utils/kalman_filter.py
def multi_predict(self, mean, covariance) -> tuple:
    """
    Run Kalman filter prediction step (Vectorized version).

    Args:
        mean (ndarray): The Nx8 dimensional mean matrix of the object states at the previous time step.
        covariance (ndarray): The Nx8x8 covariance matrix of the object states at the previous time step.

    Returns:
        (tuple[ndarray, ndarray]): Returns the mean vector and covariance matrix of the predicted state. Unobserved
            velocities are initialized to 0 mean.
    """
    std_pos = [
        self._std_weight_position * mean[:, 2],
        self._std_weight_position * mean[:, 3],
        self._std_weight_position * mean[:, 2],
        self._std_weight_position * mean[:, 3],
    ]
    std_vel = [
        self._std_weight_velocity * mean[:, 2],
        self._std_weight_velocity * mean[:, 3],
        self._std_weight_velocity * mean[:, 2],
        self._std_weight_velocity * mean[:, 3],
    ]
    sqr = np.square(np.r_[std_pos, std_vel]).T

    motion_cov = [np.diag(sqr[i]) for i in range(len(mean))]
    motion_cov = np.asarray(motion_cov)

    mean = np.dot(mean, self._motion_mat.T)
    left = np.dot(self._motion_mat, covariance).transpose((1, 0, 2))
    covariance = np.dot(left, self._motion_mat.T) + motion_cov

    return mean, covariance

predict(mean, covariance)

Exécute l'étape de prédiction du filtre de Kalman.

Paramètres :

Nom Type Description DĂ©faut
mean ndarray

Le vecteur moyen à 8 dimensions de l'état de l'objet au pas de temps précédent.

requis
covariance ndarray

La matrice de covariance de dimension 8x8 de l'état de l'objet au pas de temps précédent.

requis

Retourne :

Type Description
tuple[ndarray, ndarray]

Renvoie le vecteur moyen et la matrice de covariance de l'état prédit. Les vitesses sont initialisées à une moyenne de 0.

Code source dans ultralytics/trackers/utils/kalman_filter.py
def predict(self, mean, covariance) -> tuple:
    """
    Run Kalman filter prediction step.

    Args:
        mean (ndarray): The 8 dimensional mean vector of the object state at the previous time step.
        covariance (ndarray): The 8x8 dimensional covariance matrix of the object state at the previous time step.

    Returns:
        (tuple[ndarray, ndarray]): Returns the mean vector and covariance matrix of the predicted state. Unobserved
            velocities are initialized to 0 mean.
    """
    std_pos = [
        self._std_weight_position * mean[2],
        self._std_weight_position * mean[3],
        self._std_weight_position * mean[2],
        self._std_weight_position * mean[3],
    ]
    std_vel = [
        self._std_weight_velocity * mean[2],
        self._std_weight_velocity * mean[3],
        self._std_weight_velocity * mean[2],
        self._std_weight_velocity * mean[3],
    ]
    motion_cov = np.diag(np.square(np.r_[std_pos, std_vel]))

    mean = np.dot(mean, self._motion_mat.T)
    covariance = np.linalg.multi_dot((self._motion_mat, covariance, self._motion_mat.T)) + motion_cov

    return mean, covariance

project(mean, covariance)

Projette la distribution des Ă©tats dans l'espace de mesure.

Paramètres :

Nom Type Description DĂ©faut
mean ndarray

Le vecteur moyen de l'Ă©tat (tableau Ă  8 dimensions).

requis
covariance ndarray

La matrice de covariance de l'Ă©tat (8x8 dimensions).

requis

Retourne :

Type Description
tuple[ndarray, ndarray]

Renvoie la moyenne et la matrice de covariance projetées de l'estimation d'état donnée.

Code source dans ultralytics/trackers/utils/kalman_filter.py
def project(self, mean, covariance) -> tuple:
    """
    Project state distribution to measurement space.

    Args:
        mean (ndarray): The state's mean vector (8 dimensional array).
        covariance (ndarray): The state's covariance matrix (8x8 dimensional).

    Returns:
        (tuple[ndarray, ndarray]): Returns the projected mean and covariance matrix of the given state estimate.
    """
    std = [
        self._std_weight_position * mean[2],
        self._std_weight_position * mean[3],
        self._std_weight_position * mean[2],
        self._std_weight_position * mean[3],
    ]
    innovation_cov = np.diag(np.square(std))

    mean = np.dot(self._update_mat, mean)
    covariance = np.linalg.multi_dot((self._update_mat, covariance, self._update_mat.T))
    return mean, covariance + innovation_cov

update(mean, covariance, measurement)

Exécute l'étape de correction du filtre de Kalman.

Paramètres :

Nom Type Description DĂ©faut
mean ndarray

Le vecteur moyen de l'état prédit (8 dimensions).

requis
covariance ndarray

La matrice de covariance de l'Ă©tat (8x8 dimensions).

requis
measurement ndarray

Le vecteur de mesure à 4 dimensions (x, y, w, h), où (x, y) est la position du centre, w la largeur et h la hauteur de la boîte de délimitation. centrale, w la largeur et h la hauteur de la boîte de délimitation.

requis

Retourne :

Type Description
tuple[ndarray, ndarray]

Renvoie la distribution d'état corrigée de la mesure.

Code source dans ultralytics/trackers/utils/kalman_filter.py
def update(self, mean, covariance, measurement) -> tuple:
    """
    Run Kalman filter correction step.

    Args:
        mean (ndarray): The predicted state's mean vector (8 dimensional).
        covariance (ndarray): The state's covariance matrix (8x8 dimensional).
        measurement (ndarray): The 4 dimensional measurement vector (x, y, w, h), where (x, y) is the center
            position, w the width, and h the height of the bounding box.

    Returns:
        (tuple[ndarray, ndarray]): Returns the measurement-corrected state distribution.
    """
    return super().update(mean, covariance, measurement)





Created 2023-11-12, Updated 2024-06-02
Authors: glenn-jocher (5), Burhan-Q (1)