μ½˜ν…μΈ λ‘œ κ±΄λ„ˆλ›°κΈ°

μ°Έμ‘° ultralytics/trackers/byte_tracker.py

μ°Έκ³ 

이 νŒŒμΌμ€ https://github.com/ultralytics/ ultralytics/blob/main/ ultralytics/trackers/byte_tracker .pyμ—μ„œ 확인할 수 μžˆμŠ΅λ‹ˆλ‹€. 문제λ₯Ό λ°œκ²¬ν•˜λ©΄ ν’€ λ¦¬ν€˜μŠ€νŠΈ πŸ› οΈ 에 κΈ°μ—¬ν•˜μ—¬ 문제λ₯Ό ν•΄κ²°ν•˜λ„λ‘ λ„μ™€μ£Όμ„Έμš”. κ°μ‚¬ν•©λ‹ˆλ‹€ πŸ™!



ultralytics.trackers.byte_tracker.STrack

기지: BaseTrack

μƒνƒœ 좔정을 μœ„ν•΄ 칼만 필터링을 μ‚¬μš©ν•˜λŠ” 단일 객체 좔적 ν‘œν˜„μž…λ‹ˆλ‹€.

이 ν΄λž˜μŠ€λŠ” κ°œλ³„ νŠΈλž˜ν΄λ¦Ώμ— κ΄€ν•œ λͺ¨λ“  정보λ₯Ό μ €μž₯ν•˜κ³  칼만 필터에 κΈ°λ°˜ν•œ μƒνƒœ μ—…λ°μ΄νŠΈ 및 칼만 필터에 κΈ°λ°˜ν•œ μ˜ˆμΈ‘μ„ μˆ˜ν–‰ν•©λ‹ˆλ‹€.

속성:

이름 μœ ν˜• μ„€λͺ…
shared_kalman KalmanFilterXYAH

μ˜ˆμΈ‘μ„ μœ„ν•΄ λͺ¨λ“  STrack μΈμŠ€ν„΄μŠ€μ—μ„œ μ‚¬μš©λ˜λŠ” 곡유 칼만 ν•„ν„°μž…λ‹ˆλ‹€.

_tlwh ndarray

μ™Όμͺ½ 상단 λͺ¨μ„œλ¦¬ μ’Œν‘œμ™€ λ°”μš΄λ”© λ°•μŠ€μ˜ λ„ˆλΉ„μ™€ 높이λ₯Ό μ €μž₯ν•˜λŠ” λΉ„κ³΅κ°œ μ†μ„±μž…λ‹ˆλ‹€.

kalman_filter KalmanFilterXYAH

이 νŠΉμ • 였브젝트 νŠΈλž™μ— μ‚¬μš©λœ 칼만 ν•„ν„°μ˜ μΈμŠ€ν„΄μŠ€μž…λ‹ˆλ‹€.

mean ndarray

평균 μƒνƒœ μΆ”μ • λ²‘ν„°μž…λ‹ˆλ‹€.

covariance ndarray

μƒνƒœ μΆ”μ •μΉ˜μ˜ 곡뢄산.

is_activated bool

νŠΈλž™μ΄ ν™œμ„±ν™”λ˜μ—ˆλŠ”μ§€ μ—¬λΆ€λ₯Ό λ‚˜νƒ€λ‚΄λŠ” λΆ€μšΈ ν”Œλž˜κ·Έμž…λ‹ˆλ‹€.

score float

νŠΈλž™μ˜ 신뒰도 μ μˆ˜μž…λ‹ˆλ‹€.

tracklet_len int

νŠΈλž˜λΈ”λ ›μ˜ κΈΈμ΄μž…λ‹ˆλ‹€.

cls any

객체의 클래슀 λ ˆμ΄λΈ”μž…λ‹ˆλ‹€.

idx int

객체의 색인 λ˜λŠ” μ‹λ³„μžμž…λ‹ˆλ‹€.

frame_id int

ν˜„μž¬ ν”„λ ˆμž„ IDμž…λ‹ˆλ‹€.

start_frame int

객체가 처음 κ°μ§€λœ ν”„λ ˆμž„μž…λ‹ˆλ‹€.

λ©”μ†Œλ“œ:

이름 μ„€λͺ…
predict

칼만 ν•„ν„°λ₯Ό μ‚¬μš©ν•˜μ—¬ 객체의 λ‹€μŒ μƒνƒœλ₯Ό μ˜ˆμΈ‘ν•©λ‹ˆλ‹€.

multi_predict

μ—¬λŸ¬ νŠΈλž™μ˜ λ‹€μŒ μƒνƒœλ₯Ό μ˜ˆμΈ‘ν•˜μ„Έμš”.

multi_gmc

λ™ν˜• 행렬을 μ‚¬μš©ν•˜μ—¬ μ—¬λŸ¬ νŠΈλž™ μƒνƒœλ₯Ό μ—…λ°μ΄νŠΈν•©λ‹ˆλ‹€.

activate

μƒˆ νŠΈλž˜ν΄λ¦Ώμ„ ν™œμ„±ν™”ν•©λ‹ˆλ‹€.

re_activate

이전에 λΆ„μ‹€ν–ˆλ˜ νŠΈλž˜λΈ”λ ›μ„ λ‹€μ‹œ ν™œμ„±ν™”ν•©λ‹ˆλ‹€.

update

μΌμΉ˜ν•˜λŠ” νŠΈλž™μ˜ μƒνƒœλ₯Ό μ—…λ°μ΄νŠΈν•©λ‹ˆλ‹€.

convert_coords

λ°”μš΄λ”© λ°•μŠ€λ₯Ό x-y-μ’…νš‘λΉ„-높이 ν˜•μ‹μœΌλ‘œ λ³€ν™˜ν•©λ‹ˆλ‹€.

tlwh_to_xyah

tlwh λ°”μš΄λ”© λ°•μŠ€λ₯Ό xyah ν˜•μ‹μœΌλ‘œ λ³€ν™˜ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
class STrack(BaseTrack):
    """
    Single object tracking representation that uses Kalman filtering for state estimation.

    This class is responsible for storing all the information regarding individual tracklets and performs state updates
    and predictions based on Kalman filter.

    Attributes:
        shared_kalman (KalmanFilterXYAH): Shared Kalman filter that is used across all STrack instances for prediction.
        _tlwh (np.ndarray): Private attribute to store top-left corner coordinates and width and height of bounding box.
        kalman_filter (KalmanFilterXYAH): Instance of Kalman filter used for this particular object track.
        mean (np.ndarray): Mean state estimate vector.
        covariance (np.ndarray): Covariance of state estimate.
        is_activated (bool): Boolean flag indicating if the track has been activated.
        score (float): Confidence score of the track.
        tracklet_len (int): Length of the tracklet.
        cls (any): Class label for the object.
        idx (int): Index or identifier for the object.
        frame_id (int): Current frame ID.
        start_frame (int): Frame where the object was first detected.

    Methods:
        predict(): Predict the next state of the object using Kalman filter.
        multi_predict(stracks): Predict the next states for multiple tracks.
        multi_gmc(stracks, H): Update multiple track states using a homography matrix.
        activate(kalman_filter, frame_id): Activate a new tracklet.
        re_activate(new_track, frame_id, new_id): Reactivate a previously lost tracklet.
        update(new_track, frame_id): Update the state of a matched track.
        convert_coords(tlwh): Convert bounding box to x-y-aspect-height format.
        tlwh_to_xyah(tlwh): Convert tlwh bounding box to xyah format.
    """

    shared_kalman = KalmanFilterXYAH()

    def __init__(self, xywh, score, cls):
        """Initialize new STrack instance."""
        super().__init__()
        # xywh+idx or xywha+idx
        assert len(xywh) in [5, 6], f"expected 5 or 6 values but got {len(xywh)}"
        self._tlwh = np.asarray(xywh2ltwh(xywh[:4]), dtype=np.float32)
        self.kalman_filter = None
        self.mean, self.covariance = None, None
        self.is_activated = False

        self.score = score
        self.tracklet_len = 0
        self.cls = cls
        self.idx = xywh[-1]
        self.angle = xywh[4] if len(xywh) == 6 else None

    def predict(self):
        """Predicts mean and covariance using Kalman filter."""
        mean_state = self.mean.copy()
        if self.state != TrackState.Tracked:
            mean_state[7] = 0
        self.mean, self.covariance = self.kalman_filter.predict(mean_state, self.covariance)

    @staticmethod
    def multi_predict(stracks):
        """Perform multi-object predictive tracking using Kalman filter for given stracks."""
        if len(stracks) <= 0:
            return
        multi_mean = np.asarray([st.mean.copy() for st in stracks])
        multi_covariance = np.asarray([st.covariance for st in stracks])
        for i, st in enumerate(stracks):
            if st.state != TrackState.Tracked:
                multi_mean[i][7] = 0
        multi_mean, multi_covariance = STrack.shared_kalman.multi_predict(multi_mean, multi_covariance)
        for i, (mean, cov) in enumerate(zip(multi_mean, multi_covariance)):
            stracks[i].mean = mean
            stracks[i].covariance = cov

    @staticmethod
    def multi_gmc(stracks, H=np.eye(2, 3)):
        """Update state tracks positions and covariances using a homography matrix."""
        if len(stracks) > 0:
            multi_mean = np.asarray([st.mean.copy() for st in stracks])
            multi_covariance = np.asarray([st.covariance for st in stracks])

            R = H[:2, :2]
            R8x8 = np.kron(np.eye(4, dtype=float), R)
            t = H[:2, 2]

            for i, (mean, cov) in enumerate(zip(multi_mean, multi_covariance)):
                mean = R8x8.dot(mean)
                mean[:2] += t
                cov = R8x8.dot(cov).dot(R8x8.transpose())

                stracks[i].mean = mean
                stracks[i].covariance = cov

    def activate(self, kalman_filter, frame_id):
        """Start a new tracklet."""
        self.kalman_filter = kalman_filter
        self.track_id = self.next_id()
        self.mean, self.covariance = self.kalman_filter.initiate(self.convert_coords(self._tlwh))

        self.tracklet_len = 0
        self.state = TrackState.Tracked
        if frame_id == 1:
            self.is_activated = True
        self.frame_id = frame_id
        self.start_frame = frame_id

    def re_activate(self, new_track, frame_id, new_id=False):
        """Reactivates a previously lost track with a new detection."""
        self.mean, self.covariance = self.kalman_filter.update(
            self.mean, self.covariance, self.convert_coords(new_track.tlwh)
        )
        self.tracklet_len = 0
        self.state = TrackState.Tracked
        self.is_activated = True
        self.frame_id = frame_id
        if new_id:
            self.track_id = self.next_id()
        self.score = new_track.score
        self.cls = new_track.cls
        self.angle = new_track.angle
        self.idx = new_track.idx

    def update(self, new_track, frame_id):
        """
        Update the state of a matched track.

        Args:
            new_track (STrack): The new track containing updated information.
            frame_id (int): The ID of the current frame.
        """
        self.frame_id = frame_id
        self.tracklet_len += 1

        new_tlwh = new_track.tlwh
        self.mean, self.covariance = self.kalman_filter.update(
            self.mean, self.covariance, self.convert_coords(new_tlwh)
        )
        self.state = TrackState.Tracked
        self.is_activated = True

        self.score = new_track.score
        self.cls = new_track.cls
        self.angle = new_track.angle
        self.idx = new_track.idx

    def convert_coords(self, tlwh):
        """Convert a bounding box's top-left-width-height format to its x-y-aspect-height equivalent."""
        return self.tlwh_to_xyah(tlwh)

    @property
    def tlwh(self):
        """Get current position in bounding box format (top left x, top left y, width, height)."""
        if self.mean is None:
            return self._tlwh.copy()
        ret = self.mean[:4].copy()
        ret[2] *= ret[3]
        ret[:2] -= ret[2:] / 2
        return ret

    @property
    def xyxy(self):
        """Convert bounding box to format (min x, min y, max x, max y), i.e., (top left, bottom right)."""
        ret = self.tlwh.copy()
        ret[2:] += ret[:2]
        return ret

    @staticmethod
    def tlwh_to_xyah(tlwh):
        """Convert bounding box to format (center x, center y, aspect ratio, height), where the aspect ratio is width /
        height.
        """
        ret = np.asarray(tlwh).copy()
        ret[:2] += ret[2:] / 2
        ret[2] /= ret[3]
        return ret

    @property
    def xywh(self):
        """Get current position in bounding box format (center x, center y, width, height)."""
        ret = np.asarray(self.tlwh).copy()
        ret[:2] += ret[2:] / 2
        return ret

    @property
    def xywha(self):
        """Get current position in bounding box format (center x, center y, width, height, angle)."""
        if self.angle is None:
            LOGGER.warning("WARNING ⚠️ `angle` attr not found, returning `xywh` instead.")
            return self.xywh
        return np.concatenate([self.xywh, self.angle[None]])

    @property
    def result(self):
        """Get current tracking results."""
        coords = self.xyxy if self.angle is None else self.xywha
        return coords.tolist() + [self.track_id, self.score, self.cls, self.idx]

    def __repr__(self):
        """Return a string representation of the BYTETracker object with start and end frames and track ID."""
        return f"OT_{self.track_id}_({self.start_frame}-{self.end_frame})"

result property

ν˜„μž¬ 좔적 κ²°κ³Όλ₯Ό ν™•μΈν•˜μ„Έμš”.

tlwh property

ν˜„μž¬ μœ„μΉ˜λ₯Ό λ°”μš΄λ”© λ°•μŠ€ ν˜•μ‹(μ™Όμͺ½ 상단 x, μ™Όμͺ½ 상단 y, λ„ˆλΉ„, 높이)으둜 κ°€μ Έμ˜΅λ‹ˆλ‹€.

xywh property

ν˜„μž¬ μœ„μΉ˜λ₯Ό 경계 μƒμž ν˜•μ‹(쀑심 x, 쀑심 y, λ„ˆλΉ„, 높이)으둜 κ°€μ Έμ˜΅λ‹ˆλ‹€.

xywha property

ν˜„μž¬ μœ„μΉ˜λ₯Ό 경계 μƒμž ν˜•μ‹(쀑심 x, 쀑심 y, λ„ˆλΉ„, 높이, 각도)으둜 κ°€μ Έμ˜΅λ‹ˆλ‹€.

xyxy property

λ°”μš΄λ”© λ°•μŠ€λ₯Ό (μ΅œμ†Œ x, μ΅œμ†Œ y, μ΅œλŒ€ x, μ΅œλŒ€ y) ν˜•μ‹, 즉 (μ™Όμͺ½ μœ„, 였λ₯Έμͺ½ μ•„λž˜)둜 λ³€ν™˜ν•©λ‹ˆλ‹€.

__init__(xywh, score, cls)

μƒˆ STrack μΈμŠ€ν„΄μŠ€λ₯Ό μ΄ˆκΈ°ν™”ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def __init__(self, xywh, score, cls):
    """Initialize new STrack instance."""
    super().__init__()
    # xywh+idx or xywha+idx
    assert len(xywh) in [5, 6], f"expected 5 or 6 values but got {len(xywh)}"
    self._tlwh = np.asarray(xywh2ltwh(xywh[:4]), dtype=np.float32)
    self.kalman_filter = None
    self.mean, self.covariance = None, None
    self.is_activated = False

    self.score = score
    self.tracklet_len = 0
    self.cls = cls
    self.idx = xywh[-1]
    self.angle = xywh[4] if len(xywh) == 6 else None

__repr__()

μ‹œμž‘ 및 μ’…λ£Œ ν”„λ ˆμž„κ³Ό νŠΈλž™ IDκ°€ ν¬ν•¨λœ BYTETracker 객체의 λ¬Έμžμ—΄ ν‘œν˜„μ„ λ°˜ν™˜ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def __repr__(self):
    """Return a string representation of the BYTETracker object with start and end frames and track ID."""
    return f"OT_{self.track_id}_({self.start_frame}-{self.end_frame})"

activate(kalman_filter, frame_id)

μƒˆ νŠΈλž˜λΈ”λ ›μ„ μ‹œμž‘ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def activate(self, kalman_filter, frame_id):
    """Start a new tracklet."""
    self.kalman_filter = kalman_filter
    self.track_id = self.next_id()
    self.mean, self.covariance = self.kalman_filter.initiate(self.convert_coords(self._tlwh))

    self.tracklet_len = 0
    self.state = TrackState.Tracked
    if frame_id == 1:
        self.is_activated = True
    self.frame_id = frame_id
    self.start_frame = frame_id

convert_coords(tlwh)

λ°”μš΄λ”© λ°•μŠ€μ˜ μ™Όμͺ½ 상단-μ™Όμͺ½ λ„ˆλΉ„-높이 ν˜•μ‹μ„ ν•΄λ‹Ή x-y μ’…νš‘λΉ„-높이 ν˜•μ‹μœΌλ‘œ λ³€ν™˜ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def convert_coords(self, tlwh):
    """Convert a bounding box's top-left-width-height format to its x-y-aspect-height equivalent."""
    return self.tlwh_to_xyah(tlwh)

multi_gmc(stracks, H=np.eye(2, 3)) staticmethod

λ™ν˜• 행렬을 μ‚¬μš©ν•˜μ—¬ μƒνƒœ νŠΈλž™ μœ„μΉ˜ 및 곡뢄산을 μ—…λ°μ΄νŠΈν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
@staticmethod
def multi_gmc(stracks, H=np.eye(2, 3)):
    """Update state tracks positions and covariances using a homography matrix."""
    if len(stracks) > 0:
        multi_mean = np.asarray([st.mean.copy() for st in stracks])
        multi_covariance = np.asarray([st.covariance for st in stracks])

        R = H[:2, :2]
        R8x8 = np.kron(np.eye(4, dtype=float), R)
        t = H[:2, 2]

        for i, (mean, cov) in enumerate(zip(multi_mean, multi_covariance)):
            mean = R8x8.dot(mean)
            mean[:2] += t
            cov = R8x8.dot(cov).dot(R8x8.transpose())

            stracks[i].mean = mean
            stracks[i].covariance = cov

multi_predict(stracks) staticmethod

주어진 μŠ€νŠΈλž™μ— λŒ€ν•΄ 칼만 ν•„ν„°λ₯Ό μ‚¬μš©ν•˜μ—¬ 닀쀑 객체 예츑 좔적을 μˆ˜ν–‰ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
@staticmethod
def multi_predict(stracks):
    """Perform multi-object predictive tracking using Kalman filter for given stracks."""
    if len(stracks) <= 0:
        return
    multi_mean = np.asarray([st.mean.copy() for st in stracks])
    multi_covariance = np.asarray([st.covariance for st in stracks])
    for i, st in enumerate(stracks):
        if st.state != TrackState.Tracked:
            multi_mean[i][7] = 0
    multi_mean, multi_covariance = STrack.shared_kalman.multi_predict(multi_mean, multi_covariance)
    for i, (mean, cov) in enumerate(zip(multi_mean, multi_covariance)):
        stracks[i].mean = mean
        stracks[i].covariance = cov

predict()

칼만 ν•„ν„°λ₯Ό μ‚¬μš©ν•˜μ—¬ 평균 및 곡뢄산을 μ˜ˆμΈ‘ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def predict(self):
    """Predicts mean and covariance using Kalman filter."""
    mean_state = self.mean.copy()
    if self.state != TrackState.Tracked:
        mean_state[7] = 0
    self.mean, self.covariance = self.kalman_filter.predict(mean_state, self.covariance)

re_activate(new_track, frame_id, new_id=False)

이전에 μ†μ‹€λœ νŠΈλž™μ„ μƒˆλ‘œμš΄ νƒμ§€λ‘œ λ‹€μ‹œ ν™œμ„±ν™”ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def re_activate(self, new_track, frame_id, new_id=False):
    """Reactivates a previously lost track with a new detection."""
    self.mean, self.covariance = self.kalman_filter.update(
        self.mean, self.covariance, self.convert_coords(new_track.tlwh)
    )
    self.tracklet_len = 0
    self.state = TrackState.Tracked
    self.is_activated = True
    self.frame_id = frame_id
    if new_id:
        self.track_id = self.next_id()
    self.score = new_track.score
    self.cls = new_track.cls
    self.angle = new_track.angle
    self.idx = new_track.idx

tlwh_to_xyah(tlwh) staticmethod

경계 μƒμžλ₯Ό ν˜•μ‹(쀑심 x, 쀑심 y, κ°€λ‘œ μ„Έλ‘œ λΉ„μœ¨, 높이)으둜 λ³€ν™˜ν•©λ‹ˆλ‹€. μ—¬κΈ°μ„œ κ°€λ‘œ μ„Έλ‘œ λΉ„μœ¨μ€ λ„ˆλΉ„ / λ†’μ΄μž…λ‹ˆλ‹€. λ†’μ΄μž…λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
@staticmethod
def tlwh_to_xyah(tlwh):
    """Convert bounding box to format (center x, center y, aspect ratio, height), where the aspect ratio is width /
    height.
    """
    ret = np.asarray(tlwh).copy()
    ret[:2] += ret[2:] / 2
    ret[2] /= ret[3]
    return ret

update(new_track, frame_id)

μΌμΉ˜ν•˜λŠ” νŠΈλž™μ˜ μƒνƒœλ₯Ό μ—…λ°μ΄νŠΈν•©λ‹ˆλ‹€.

λ§€κ°œλ³€μˆ˜:

이름 μœ ν˜• μ„€λͺ… κΈ°λ³Έκ°’
new_track STrack

μ—…λ°μ΄νŠΈλœ 정보가 ν¬ν•¨λœ μƒˆ νŠΈλž™μž…λ‹ˆλ‹€.

ν•„μˆ˜
frame_id int

ν˜„μž¬ ν”„λ ˆμž„μ˜ IDμž…λ‹ˆλ‹€.

ν•„μˆ˜
의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def update(self, new_track, frame_id):
    """
    Update the state of a matched track.

    Args:
        new_track (STrack): The new track containing updated information.
        frame_id (int): The ID of the current frame.
    """
    self.frame_id = frame_id
    self.tracklet_len += 1

    new_tlwh = new_track.tlwh
    self.mean, self.covariance = self.kalman_filter.update(
        self.mean, self.covariance, self.convert_coords(new_tlwh)
    )
    self.state = TrackState.Tracked
    self.is_activated = True

    self.score = new_track.score
    self.cls = new_track.cls
    self.angle = new_track.angle
    self.idx = new_track.idx



ultralytics.trackers.byte_tracker.BYTETracker

BYTETracker: 개체 감지 및 좔적을 μœ„ν•΄ YOLOv8 μœ„μ— κ΅¬μΆ•λœ 좔적 μ•Œκ³ λ¦¬μ¦˜μž…λ‹ˆλ‹€.

이 ν΄λž˜μŠ€λŠ” λΉ„λ””μ˜€ μ‹œν€€μŠ€μ—μ„œ κ°μ§€λœ 였브젝트의 νŠΈλž™μ„ μ΄ˆκΈ°ν™”, μ—…λ°μ΄νŠΈ, κ΄€λ¦¬ν•˜λŠ” 역할을 ν•©λ‹ˆλ‹€. μ‹œν€€μŠ€μž…λ‹ˆλ‹€. 이 ν΄λž˜μŠ€λŠ” ν”„λ ˆμž„μ— 걸쳐 좔적, 손싀, 제거된 νŠΈλž™μ˜ μƒνƒœλ₯Ό μœ μ§€ν•˜κ³ , 칼만 필터링을 ν™œμš©ν•˜μ—¬ μƒˆλ‘œμš΄ 객체 μœ„μΉ˜λ₯Ό μ˜ˆμΈ‘ν•˜κ³  데이터 연결을 μˆ˜ν–‰ν•©λ‹ˆλ‹€.

속성:

이름 μœ ν˜• μ„€λͺ…
tracked_stracks list[STrack]

μ„±κ³΅μ μœΌλ‘œ ν™œμ„±ν™”λœ νŠΈλž™ λͺ©λ‘μž…λ‹ˆλ‹€.

lost_stracks list[STrack]

μ†μ‹€λœ νŠΈλž™ λͺ©λ‘.

removed_stracks list[STrack]

제거된 νŠΈλž™ λͺ©λ‘μž…λ‹ˆλ‹€.

frame_id int

ν˜„μž¬ ν”„λ ˆμž„ IDμž…λ‹ˆλ‹€.

args namespace

λͺ…령쀄 인수.

max_time_lost int

νŠΈλž™μ„ '손싀'둜 κ°„μ£Όν•  수 μžˆλŠ” μ΅œλŒ€ ν”„λ ˆμž„μž…λ‹ˆλ‹€.

kalman_filter object

칼만 ν•„ν„° κ°μ²΄μž…λ‹ˆλ‹€.

λ©”μ†Œλ“œ:

이름 μ„€λͺ…
update

μƒˆλ‘œμš΄ 감지 κΈ°λŠ₯으둜 객체 좔적기λ₯Ό μ—…λ°μ΄νŠΈν•©λ‹ˆλ‹€.

get_kalmanfilter

λ°”μš΄λ”© λ°•μŠ€ 좔적을 μœ„ν•œ 칼만 ν•„ν„° 객체λ₯Ό λ°˜ν™˜ν•©λ‹ˆλ‹€.

init_track

감지λ₯Ό 톡해 객체 좔적을 μ΄ˆκΈ°ν™”ν•©λ‹ˆλ‹€.

get_dists

νŠΈλž™κ³Ό 탐지 μ‚¬μ΄μ˜ 거리λ₯Ό κ³„μ‚°ν•©λ‹ˆλ‹€.

multi_predict

νŠΈλž™μ˜ μœ„μΉ˜λ₯Ό μ˜ˆμΈ‘ν•©λ‹ˆλ‹€.

reset_id

STrack의 ID μΉ΄μš΄ν„°λ₯Ό μ΄ˆκΈ°ν™”ν•©λ‹ˆλ‹€.

joint_stracks

두 개의 μŠ€νŠΈλž™ λͺ©λ‘μ„ κ²°ν•©ν•©λ‹ˆλ‹€.

sub_stracks

첫 번째 λͺ©λ‘μ—μ„œ 두 번째 λͺ©λ‘μ— μžˆλŠ” μŠ€νŠΈλž™μ„ ν•„ν„°λ§ν•©λ‹ˆλ‹€.

remove_duplicate_stracks

IOUλ₯Ό 기반으둜 μ€‘λ³΅λœ μŠ€νŠΈλž™μ„ μ œκ±°ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
class BYTETracker:
    """
    BYTETracker: A tracking algorithm built on top of YOLOv8 for object detection and tracking.

    The class is responsible for initializing, updating, and managing the tracks for detected objects in a video
    sequence. It maintains the state of tracked, lost, and removed tracks over frames, utilizes Kalman filtering for
    predicting the new object locations, and performs data association.

    Attributes:
        tracked_stracks (list[STrack]): List of successfully activated tracks.
        lost_stracks (list[STrack]): List of lost tracks.
        removed_stracks (list[STrack]): List of removed tracks.
        frame_id (int): The current frame ID.
        args (namespace): Command-line arguments.
        max_time_lost (int): The maximum frames for a track to be considered as 'lost'.
        kalman_filter (object): Kalman Filter object.

    Methods:
        update(results, img=None): Updates object tracker with new detections.
        get_kalmanfilter(): Returns a Kalman filter object for tracking bounding boxes.
        init_track(dets, scores, cls, img=None): Initialize object tracking with detections.
        get_dists(tracks, detections): Calculates the distance between tracks and detections.
        multi_predict(tracks): Predicts the location of tracks.
        reset_id(): Resets the ID counter of STrack.
        joint_stracks(tlista, tlistb): Combines two lists of stracks.
        sub_stracks(tlista, tlistb): Filters out the stracks present in the second list from the first list.
        remove_duplicate_stracks(stracksa, stracksb): Removes duplicate stracks based on IOU.
    """

    def __init__(self, args, frame_rate=30):
        """Initialize a YOLOv8 object to track objects with given arguments and frame rate."""
        self.tracked_stracks = []  # type: list[STrack]
        self.lost_stracks = []  # type: list[STrack]
        self.removed_stracks = []  # type: list[STrack]

        self.frame_id = 0
        self.args = args
        self.max_time_lost = int(frame_rate / 30.0 * args.track_buffer)
        self.kalman_filter = self.get_kalmanfilter()
        self.reset_id()

    def update(self, results, img=None):
        """Updates object tracker with new detections and returns tracked object bounding boxes."""
        self.frame_id += 1
        activated_stracks = []
        refind_stracks = []
        lost_stracks = []
        removed_stracks = []

        scores = results.conf
        bboxes = results.xywhr if hasattr(results, "xywhr") else results.xywh
        # Add index
        bboxes = np.concatenate([bboxes, np.arange(len(bboxes)).reshape(-1, 1)], axis=-1)
        cls = results.cls

        remain_inds = scores > self.args.track_high_thresh
        inds_low = scores > self.args.track_low_thresh
        inds_high = scores < self.args.track_high_thresh

        inds_second = np.logical_and(inds_low, inds_high)
        dets_second = bboxes[inds_second]
        dets = bboxes[remain_inds]
        scores_keep = scores[remain_inds]
        scores_second = scores[inds_second]
        cls_keep = cls[remain_inds]
        cls_second = cls[inds_second]

        detections = self.init_track(dets, scores_keep, cls_keep, img)
        # Add newly detected tracklets to tracked_stracks
        unconfirmed = []
        tracked_stracks = []  # type: list[STrack]
        for track in self.tracked_stracks:
            if not track.is_activated:
                unconfirmed.append(track)
            else:
                tracked_stracks.append(track)
        # Step 2: First association, with high score detection boxes
        strack_pool = self.joint_stracks(tracked_stracks, self.lost_stracks)
        # Predict the current location with KF
        self.multi_predict(strack_pool)
        if hasattr(self, "gmc") and img is not None:
            warp = self.gmc.apply(img, dets)
            STrack.multi_gmc(strack_pool, warp)
            STrack.multi_gmc(unconfirmed, warp)

        dists = self.get_dists(strack_pool, detections)
        matches, u_track, u_detection = matching.linear_assignment(dists, thresh=self.args.match_thresh)

        for itracked, idet in matches:
            track = strack_pool[itracked]
            det = detections[idet]
            if track.state == TrackState.Tracked:
                track.update(det, self.frame_id)
                activated_stracks.append(track)
            else:
                track.re_activate(det, self.frame_id, new_id=False)
                refind_stracks.append(track)
        # Step 3: Second association, with low score detection boxes association the untrack to the low score detections
        detections_second = self.init_track(dets_second, scores_second, cls_second, img)
        r_tracked_stracks = [strack_pool[i] for i in u_track if strack_pool[i].state == TrackState.Tracked]
        # TODO
        dists = matching.iou_distance(r_tracked_stracks, detections_second)
        matches, u_track, u_detection_second = matching.linear_assignment(dists, thresh=0.5)
        for itracked, idet in matches:
            track = r_tracked_stracks[itracked]
            det = detections_second[idet]
            if track.state == TrackState.Tracked:
                track.update(det, self.frame_id)
                activated_stracks.append(track)
            else:
                track.re_activate(det, self.frame_id, new_id=False)
                refind_stracks.append(track)

        for it in u_track:
            track = r_tracked_stracks[it]
            if track.state != TrackState.Lost:
                track.mark_lost()
                lost_stracks.append(track)
        # Deal with unconfirmed tracks, usually tracks with only one beginning frame
        detections = [detections[i] for i in u_detection]
        dists = self.get_dists(unconfirmed, detections)
        matches, u_unconfirmed, u_detection = matching.linear_assignment(dists, thresh=0.7)
        for itracked, idet in matches:
            unconfirmed[itracked].update(detections[idet], self.frame_id)
            activated_stracks.append(unconfirmed[itracked])
        for it in u_unconfirmed:
            track = unconfirmed[it]
            track.mark_removed()
            removed_stracks.append(track)
        # Step 4: Init new stracks
        for inew in u_detection:
            track = detections[inew]
            if track.score < self.args.new_track_thresh:
                continue
            track.activate(self.kalman_filter, self.frame_id)
            activated_stracks.append(track)
        # Step 5: Update state
        for track in self.lost_stracks:
            if self.frame_id - track.end_frame > self.max_time_lost:
                track.mark_removed()
                removed_stracks.append(track)

        self.tracked_stracks = [t for t in self.tracked_stracks if t.state == TrackState.Tracked]
        self.tracked_stracks = self.joint_stracks(self.tracked_stracks, activated_stracks)
        self.tracked_stracks = self.joint_stracks(self.tracked_stracks, refind_stracks)
        self.lost_stracks = self.sub_stracks(self.lost_stracks, self.tracked_stracks)
        self.lost_stracks.extend(lost_stracks)
        self.lost_stracks = self.sub_stracks(self.lost_stracks, self.removed_stracks)
        self.tracked_stracks, self.lost_stracks = self.remove_duplicate_stracks(self.tracked_stracks, self.lost_stracks)
        self.removed_stracks.extend(removed_stracks)
        if len(self.removed_stracks) > 1000:
            self.removed_stracks = self.removed_stracks[-999:]  # clip remove stracks to 1000 maximum

        return np.asarray([x.result for x in self.tracked_stracks if x.is_activated], dtype=np.float32)

    def get_kalmanfilter(self):
        """Returns a Kalman filter object for tracking bounding boxes."""
        return KalmanFilterXYAH()

    def init_track(self, dets, scores, cls, img=None):
        """Initialize object tracking with detections and scores using STrack algorithm."""
        return [STrack(xyxy, s, c) for (xyxy, s, c) in zip(dets, scores, cls)] if len(dets) else []  # detections

    def get_dists(self, tracks, detections):
        """Calculates the distance between tracks and detections using IOU and fuses scores."""
        dists = matching.iou_distance(tracks, detections)
        # TODO: mot20
        # if not self.args.mot20:
        dists = matching.fuse_score(dists, detections)
        return dists

    def multi_predict(self, tracks):
        """Returns the predicted tracks using the YOLOv8 network."""
        STrack.multi_predict(tracks)

    @staticmethod
    def reset_id():
        """Resets the ID counter of STrack."""
        STrack.reset_id()

    def reset(self):
        """Reset tracker."""
        self.tracked_stracks = []  # type: list[STrack]
        self.lost_stracks = []  # type: list[STrack]
        self.removed_stracks = []  # type: list[STrack]
        self.frame_id = 0
        self.kalman_filter = self.get_kalmanfilter()
        self.reset_id()

    @staticmethod
    def joint_stracks(tlista, tlistb):
        """Combine two lists of stracks into a single one."""
        exists = {}
        res = []
        for t in tlista:
            exists[t.track_id] = 1
            res.append(t)
        for t in tlistb:
            tid = t.track_id
            if not exists.get(tid, 0):
                exists[tid] = 1
                res.append(t)
        return res

    @staticmethod
    def sub_stracks(tlista, tlistb):
        """DEPRECATED CODE in https://github.com/ultralytics/ultralytics/pull/1890/
        stracks = {t.track_id: t for t in tlista}
        for t in tlistb:
            tid = t.track_id
            if stracks.get(tid, 0):
                del stracks[tid]
        return list(stracks.values())
        """
        track_ids_b = {t.track_id for t in tlistb}
        return [t for t in tlista if t.track_id not in track_ids_b]

    @staticmethod
    def remove_duplicate_stracks(stracksa, stracksb):
        """Remove duplicate stracks with non-maximum IOU distance."""
        pdist = matching.iou_distance(stracksa, stracksb)
        pairs = np.where(pdist < 0.15)
        dupa, dupb = [], []
        for p, q in zip(*pairs):
            timep = stracksa[p].frame_id - stracksa[p].start_frame
            timeq = stracksb[q].frame_id - stracksb[q].start_frame
            if timep > timeq:
                dupb.append(q)
            else:
                dupa.append(p)
        resa = [t for i, t in enumerate(stracksa) if i not in dupa]
        resb = [t for i, t in enumerate(stracksb) if i not in dupb]
        return resa, resb

__init__(args, frame_rate=30)

YOLOv8 객체λ₯Ό μ΄ˆκΈ°ν™”ν•˜μ—¬ 주어진 μΈμˆ˜μ™€ ν”„λ ˆμž„ μ†λ„λ‘œ 객체λ₯Ό μΆ”μ ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def __init__(self, args, frame_rate=30):
    """Initialize a YOLOv8 object to track objects with given arguments and frame rate."""
    self.tracked_stracks = []  # type: list[STrack]
    self.lost_stracks = []  # type: list[STrack]
    self.removed_stracks = []  # type: list[STrack]

    self.frame_id = 0
    self.args = args
    self.max_time_lost = int(frame_rate / 30.0 * args.track_buffer)
    self.kalman_filter = self.get_kalmanfilter()
    self.reset_id()

get_dists(tracks, detections)

IOUλ₯Ό μ‚¬μš©ν•˜μ—¬ νŠΈλž™κ³Ό 탐지 μ‚¬μ΄μ˜ 거리λ₯Ό κ³„μ‚°ν•˜κ³  점수λ₯Ό μœ΅ν•©ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def get_dists(self, tracks, detections):
    """Calculates the distance between tracks and detections using IOU and fuses scores."""
    dists = matching.iou_distance(tracks, detections)
    # TODO: mot20
    # if not self.args.mot20:
    dists = matching.fuse_score(dists, detections)
    return dists

get_kalmanfilter()

λ°”μš΄λ”© λ°•μŠ€ 좔적을 μœ„ν•œ 칼만 ν•„ν„° 객체λ₯Ό λ°˜ν™˜ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def get_kalmanfilter(self):
    """Returns a Kalman filter object for tracking bounding boxes."""
    return KalmanFilterXYAH()

init_track(dets, scores, cls, img=None)

STrack μ•Œκ³ λ¦¬μ¦˜μ„ μ‚¬μš©ν•˜μ—¬ 감지 및 점수둜 객체 좔적을 μ΄ˆκΈ°ν™”ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def init_track(self, dets, scores, cls, img=None):
    """Initialize object tracking with detections and scores using STrack algorithm."""
    return [STrack(xyxy, s, c) for (xyxy, s, c) in zip(dets, scores, cls)] if len(dets) else []  # detections

joint_stracks(tlista, tlistb) staticmethod

두 개의 μŠ€νŠΈλž™ λͺ©λ‘μ„ ν•˜λ‚˜μ˜ λͺ©λ‘μœΌλ‘œ ν•©μΉ©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
@staticmethod
def joint_stracks(tlista, tlistb):
    """Combine two lists of stracks into a single one."""
    exists = {}
    res = []
    for t in tlista:
        exists[t.track_id] = 1
        res.append(t)
    for t in tlistb:
        tid = t.track_id
        if not exists.get(tid, 0):
            exists[tid] = 1
            res.append(t)
    return res

multi_predict(tracks)

YOLOv8 λ„€νŠΈμ›Œν¬λ₯Ό μ‚¬μš©ν•˜μ—¬ 예츑된 νŠΈλž™μ„ λ°˜ν™˜ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def multi_predict(self, tracks):
    """Returns the predicted tracks using the YOLOv8 network."""
    STrack.multi_predict(tracks)

remove_duplicate_stracks(stracksa, stracksb) staticmethod

μ΅œλŒ€ IOU 거리가 μ•„λ‹Œ 쀑볡 μŠ€νŠΈλž™μ„ μ œκ±°ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
@staticmethod
def remove_duplicate_stracks(stracksa, stracksb):
    """Remove duplicate stracks with non-maximum IOU distance."""
    pdist = matching.iou_distance(stracksa, stracksb)
    pairs = np.where(pdist < 0.15)
    dupa, dupb = [], []
    for p, q in zip(*pairs):
        timep = stracksa[p].frame_id - stracksa[p].start_frame
        timeq = stracksb[q].frame_id - stracksb[q].start_frame
        if timep > timeq:
            dupb.append(q)
        else:
            dupa.append(p)
    resa = [t for i, t in enumerate(stracksa) if i not in dupa]
    resb = [t for i, t in enumerate(stracksb) if i not in dupb]
    return resa, resb

reset()

트래컀λ₯Ό μž¬μ„€μ •ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def reset(self):
    """Reset tracker."""
    self.tracked_stracks = []  # type: list[STrack]
    self.lost_stracks = []  # type: list[STrack]
    self.removed_stracks = []  # type: list[STrack]
    self.frame_id = 0
    self.kalman_filter = self.get_kalmanfilter()
    self.reset_id()

reset_id() staticmethod

STrack의 ID μΉ΄μš΄ν„°λ₯Ό μ΄ˆκΈ°ν™”ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
@staticmethod
def reset_id():
    """Resets the ID counter of STrack."""
    STrack.reset_id()

sub_stracks(tlista, tlistb) staticmethod

DEPRECATED CODE in https://github.com/ultralytics/ultralytics/pull/1890/ stracks = {t.track_id: t for t in tlista} for t in tlistb: tid = t.track_id if stracks.get(tid, 0): del stracks[tid] return list(stracks.values())

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
@staticmethod
def sub_stracks(tlista, tlistb):
    """DEPRECATED CODE in https://github.com/ultralytics/ultralytics/pull/1890/
    stracks = {t.track_id: t for t in tlista}
    for t in tlistb:
        tid = t.track_id
        if stracks.get(tid, 0):
            del stracks[tid]
    return list(stracks.values())
    """
    track_ids_b = {t.track_id for t in tlistb}
    return [t for t in tlista if t.track_id not in track_ids_b]

update(results, img=None)

μƒˆλ‘œμš΄ 감지 κΈ°λŠ₯으둜 객체 좔적기λ₯Ό μ—…λ°μ΄νŠΈν•˜κ³  μΆ”μ λœ 객체 경계 μƒμžλ₯Ό λ°˜ν™˜ν•©λ‹ˆλ‹€.

의 μ†ŒμŠ€ μ½”λ“œ ultralytics/trackers/byte_tracker.py
def update(self, results, img=None):
    """Updates object tracker with new detections and returns tracked object bounding boxes."""
    self.frame_id += 1
    activated_stracks = []
    refind_stracks = []
    lost_stracks = []
    removed_stracks = []

    scores = results.conf
    bboxes = results.xywhr if hasattr(results, "xywhr") else results.xywh
    # Add index
    bboxes = np.concatenate([bboxes, np.arange(len(bboxes)).reshape(-1, 1)], axis=-1)
    cls = results.cls

    remain_inds = scores > self.args.track_high_thresh
    inds_low = scores > self.args.track_low_thresh
    inds_high = scores < self.args.track_high_thresh

    inds_second = np.logical_and(inds_low, inds_high)
    dets_second = bboxes[inds_second]
    dets = bboxes[remain_inds]
    scores_keep = scores[remain_inds]
    scores_second = scores[inds_second]
    cls_keep = cls[remain_inds]
    cls_second = cls[inds_second]

    detections = self.init_track(dets, scores_keep, cls_keep, img)
    # Add newly detected tracklets to tracked_stracks
    unconfirmed = []
    tracked_stracks = []  # type: list[STrack]
    for track in self.tracked_stracks:
        if not track.is_activated:
            unconfirmed.append(track)
        else:
            tracked_stracks.append(track)
    # Step 2: First association, with high score detection boxes
    strack_pool = self.joint_stracks(tracked_stracks, self.lost_stracks)
    # Predict the current location with KF
    self.multi_predict(strack_pool)
    if hasattr(self, "gmc") and img is not None:
        warp = self.gmc.apply(img, dets)
        STrack.multi_gmc(strack_pool, warp)
        STrack.multi_gmc(unconfirmed, warp)

    dists = self.get_dists(strack_pool, detections)
    matches, u_track, u_detection = matching.linear_assignment(dists, thresh=self.args.match_thresh)

    for itracked, idet in matches:
        track = strack_pool[itracked]
        det = detections[idet]
        if track.state == TrackState.Tracked:
            track.update(det, self.frame_id)
            activated_stracks.append(track)
        else:
            track.re_activate(det, self.frame_id, new_id=False)
            refind_stracks.append(track)
    # Step 3: Second association, with low score detection boxes association the untrack to the low score detections
    detections_second = self.init_track(dets_second, scores_second, cls_second, img)
    r_tracked_stracks = [strack_pool[i] for i in u_track if strack_pool[i].state == TrackState.Tracked]
    # TODO
    dists = matching.iou_distance(r_tracked_stracks, detections_second)
    matches, u_track, u_detection_second = matching.linear_assignment(dists, thresh=0.5)
    for itracked, idet in matches:
        track = r_tracked_stracks[itracked]
        det = detections_second[idet]
        if track.state == TrackState.Tracked:
            track.update(det, self.frame_id)
            activated_stracks.append(track)
        else:
            track.re_activate(det, self.frame_id, new_id=False)
            refind_stracks.append(track)

    for it in u_track:
        track = r_tracked_stracks[it]
        if track.state != TrackState.Lost:
            track.mark_lost()
            lost_stracks.append(track)
    # Deal with unconfirmed tracks, usually tracks with only one beginning frame
    detections = [detections[i] for i in u_detection]
    dists = self.get_dists(unconfirmed, detections)
    matches, u_unconfirmed, u_detection = matching.linear_assignment(dists, thresh=0.7)
    for itracked, idet in matches:
        unconfirmed[itracked].update(detections[idet], self.frame_id)
        activated_stracks.append(unconfirmed[itracked])
    for it in u_unconfirmed:
        track = unconfirmed[it]
        track.mark_removed()
        removed_stracks.append(track)
    # Step 4: Init new stracks
    for inew in u_detection:
        track = detections[inew]
        if track.score < self.args.new_track_thresh:
            continue
        track.activate(self.kalman_filter, self.frame_id)
        activated_stracks.append(track)
    # Step 5: Update state
    for track in self.lost_stracks:
        if self.frame_id - track.end_frame > self.max_time_lost:
            track.mark_removed()
            removed_stracks.append(track)

    self.tracked_stracks = [t for t in self.tracked_stracks if t.state == TrackState.Tracked]
    self.tracked_stracks = self.joint_stracks(self.tracked_stracks, activated_stracks)
    self.tracked_stracks = self.joint_stracks(self.tracked_stracks, refind_stracks)
    self.lost_stracks = self.sub_stracks(self.lost_stracks, self.tracked_stracks)
    self.lost_stracks.extend(lost_stracks)
    self.lost_stracks = self.sub_stracks(self.lost_stracks, self.removed_stracks)
    self.tracked_stracks, self.lost_stracks = self.remove_duplicate_stracks(self.tracked_stracks, self.lost_stracks)
    self.removed_stracks.extend(removed_stracks)
    if len(self.removed_stracks) > 1000:
        self.removed_stracks = self.removed_stracks[-999:]  # clip remove stracks to 1000 maximum

    return np.asarray([x.result for x in self.tracked_stracks if x.is_activated], dtype=np.float32)





2023-11-12 생성, 2023-11-25 μ—…λ°μ΄νŠΈλ¨
μž‘μ„±μž: glenn-jocher (3), Laughing-q (1)