Skip to content

Boxes Utils

move_boxes

supervision.detection.utils.boxes.move_boxes(xyxy: npt.NDArray[np.number], offset: npt.NDArray[np.integer]) -> npt.NDArray[np.number]

Move axis-aligned boxes by an (x, y) offset.

Parameters:

Name Type Description Default

xyxy

NDArray[number]

An array of shape (n, 4) containing the bounding boxes coordinates in format [x1, y1, x2, y2]

 required

offset

NDArray[integer]

An array of shape (2,) containing offset values in format is [dx, dy].

 required

Returns:

Type Description
NDArray[number]

Repositioned bounding boxes. Integer inputs produce int64 output when
NDArray[number]

the translated coordinates fit; otherwise, they produce float64 output.
NDArray[number]

Floating-point inputs preserve NumPy's existing addition behavior.

Examples:

>>> import numpy as np
>>> import supervision as sv
>>> xyxy = np.array([
...     [10, 10, 20, 20],
...     [30, 30, 40, 40]
... ])
>>> offset = np.array([5, 5])
>>> sv.move_boxes(xyxy=xyxy, offset=offset)
array([[15, 15, 25, 25],
       [35, 35, 45, 45]])
Source code in src/supervision/detection/utils/boxes.py 
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
def move_boxes(
    xyxy: npt.NDArray[np.number], offset: npt.NDArray[np.integer]
) -> npt.NDArray[np.number]:
    """Move axis-aligned boxes by an `(x, y)` offset.

    Args:
        xyxy: An array of shape `(n, 4)` containing the
            bounding boxes coordinates in format `[x1, y1, x2, y2]`
        offset: An array of shape `(2,)` containing offset values in format
            is `[dx, dy]`.

    Returns:
        Repositioned bounding boxes. Integer inputs produce `int64` output when
        the translated coordinates fit; otherwise, they produce `float64` output.
        Floating-point inputs preserve NumPy's existing addition behavior.

    Examples:
        ```pycon
        >>> import numpy as np
        >>> import supervision as sv
        >>> xyxy = np.array([
        ...     [10, 10, 20, 20],
        ...     [30, 30, 40, 40]
        ... ])
        >>> offset = np.array([5, 5])
        >>> sv.move_boxes(xyxy=xyxy, offset=offset)
        array([[15, 15, 25, 25],
               [35, 35, 45, 45]])

        ```
    """
    offset = np.asarray(offset)
    return _add_integer_offset(xyxy, np.hstack([offset, offset]))

scale_boxes

supervision.detection.utils.boxes.scale_boxes(xyxy: npt.NDArray[np.number], factor: float) -> npt.NDArray[np.floating]

Scale the dimensions of bounding boxes.

Parameters:

Name Type Description Default

xyxy

NDArray[number]

An integer or floating-point array of shape (n, 4) containing bounding box coordinates in format [x1, y1, x2, y2].

 required

factor

float

A float value representing the factor by which the box dimensions are scaled. A factor greater than 1 enlarges the boxes, while a factor less than 1 shrinks them.

 required

Returns:

Type Description
NDArray[floating]

Scaled bounding boxes. Integer input arrays produce float64 output;
NDArray[floating]

floating-point inputs preserve their existing dtype.

Examples:

>>> import numpy as np
>>> import supervision as sv
>>> xyxy = np.array([
...     [10, 10, 20, 20],
...     [30, 30, 40, 40]
... ])
>>> sv.scale_boxes(xyxy=xyxy, factor=1.5)
array([[ 7.5,  7.5, 22.5, 22.5],
       [27.5, 27.5, 42.5, 42.5]])
Source code in src/supervision/detection/utils/boxes.py 
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
def scale_boxes(
    xyxy: npt.NDArray[np.number], factor: float
) -> npt.NDArray[np.floating]:
    """Scale the dimensions of bounding boxes.

    Args:
        xyxy: An integer or floating-point array of shape `(n, 4)` containing
            bounding box coordinates in format `[x1, y1, x2, y2]`.
        factor: A float value representing the factor by which the box
            dimensions are scaled. A factor greater than 1 enlarges the boxes, while a
            factor less than 1 shrinks them.

    Returns:
        Scaled bounding boxes. Integer input arrays produce `float64` output;
        floating-point inputs preserve their existing dtype.

    Examples:
        ```pycon
        >>> import numpy as np
        >>> import supervision as sv
        >>> xyxy = np.array([
        ...     [10, 10, 20, 20],
        ...     [30, 30, 40, 40]
        ... ])
        >>> sv.scale_boxes(xyxy=xyxy, factor=1.5)
        array([[ 7.5,  7.5, 22.5, 22.5],
               [27.5, 27.5, 42.5, 42.5]])

        ```
    """
    if np.issubdtype(xyxy.dtype, np.integer):
        exact_integer_limit = 2**52
        is_safe_to_cast = np.all(
            xyxy <= exact_integer_limit
            if np.issubdtype(xyxy.dtype, np.unsignedinteger)
            else (xyxy >= -exact_integer_limit) & (xyxy <= exact_integer_limit)
        )
        if is_safe_to_cast:
            xyxy = xyxy.astype(np.float64)
        elif np.isfinite(factor):
            # Preserve 64-bit integer precision through both final-corner
            # expressions. Converting the center first can change the final
            # float64 rounding near 2**53.
            factor_numerator, factor_denominator = factor.as_integer_ratio()
            integer_coordinates = xyxy.astype(object)
            coordinate_sum = integer_coordinates[:, :2] + integer_coordinates[:, 2:]
            coordinate_size = integer_coordinates[:, 2:] - integer_coordinates[:, :2]
            denominator = 2 * factor_denominator
            lower = np.asarray(
                (
                    coordinate_sum * factor_denominator
                    - coordinate_size * factor_numerator
                )
                / denominator,
                dtype=np.float64,
            )
            upper = np.asarray(
                (
                    coordinate_sum * factor_denominator
                    + coordinate_size * factor_numerator
                )
                / denominator,
                dtype=np.float64,
            )
            return cast(
                npt.NDArray[np.floating], np.concatenate((lower, upper), axis=1)
            )
        else:
            xyxy = xyxy.astype(np.float64)

    centers = cast(npt.NDArray[np.floating], (xyxy[:, :2] + xyxy[:, 2:]) / 2)
    new_sizes = cast(npt.NDArray[np.floating], (xyxy[:, 2:] - xyxy[:, :2]) * factor)
    return cast(
        npt.NDArray[np.floating],
        np.concatenate((centers - new_sizes / 2, centers + new_sizes / 2), axis=1),
    )

clip_boxes

supervision.detection.utils.boxes.clip_boxes(xyxy: npt.NDArray[np.number], resolution_wh: tuple[int, int]) -> npt.NDArray[np.number]

Clips bounding boxes coordinates to fit within the frame resolution.

Parameters:

Name Type Description Default

xyxy

NDArray[number]

A numpy array of shape (N, 4) where each row corresponds to a bounding box in the format (x_min, y_min, x_max, y_max).

 required

resolution_wh

tuple[int, int]

A tuple of the form (width, height) representing the resolution of the frame.

 required

Returns:

Type Description
NDArray[number]

A numpy array of shape (N, 4) where each row corresponds to a bounding box with coordinates clipped to fit within the frame resolution.

Examples:

>>> import numpy as np
>>> import supervision as sv
>>> xyxy = np.array([
...     [10, 20, 300, 200],
...     [15, 25, 350, 450],
...     [-10, -20, 30, 40]
... ])
>>> sv.clip_boxes(xyxy=xyxy, resolution_wh=(320, 240))
array([[ 10,  20, 300, 200],
       [ 15,  25, 320, 240],
       [  0,   0,  30,  40]])
Source code in src/supervision/detection/utils/boxes.py 
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
def clip_boxes(
    xyxy: npt.NDArray[np.number],
    resolution_wh: tuple[int, int],
) -> npt.NDArray[np.number]:
    """Clips bounding boxes coordinates to fit within the frame resolution.

    Args:
        xyxy: A numpy array of shape `(N, 4)` where each
            row corresponds to a bounding box in
            the format `(x_min, y_min, x_max, y_max)`.
        resolution_wh: A tuple of the form
            `(width, height)` representing the resolution of the frame.

    Returns:
        A numpy array of shape `(N, 4)` where each row
            corresponds to a bounding box with coordinates clipped to fit
            within the frame resolution.

    Examples:
        ```pycon
        >>> import numpy as np
        >>> import supervision as sv
        >>> xyxy = np.array([
        ...     [10, 20, 300, 200],
        ...     [15, 25, 350, 450],
        ...     [-10, -20, 30, 40]
        ... ])
        >>> sv.clip_boxes(xyxy=xyxy, resolution_wh=(320, 240))
        array([[ 10,  20, 300, 200],
               [ 15,  25, 320, 240],
               [  0,   0,  30,  40]])

        ```
    """
    result: npt.NDArray[np.number] = np.copy(xyxy)
    width, height = resolution_wh
    result[:, [0, 2]] = result[:, [0, 2]].clip(0, width)
    result[:, [1, 3]] = result[:, [1, 3]].clip(0, height)
    return result

pad_boxes

supervision.detection.utils.boxes.pad_boxes(xyxy: npt.NDArray[np.number], px: int, py: int | None = None) -> npt.NDArray[np.number]

Pads bounding boxes coordinates with a constant padding.

Parameters:

Name Type Description Default

xyxy

NDArray[number]

A numpy array of shape (N, 4) where each row corresponds to a bounding box in the format (x_min, y_min, x_max, y_max).

 required

px

int

The padding value to be added to both the left and right sides of each bounding box.

 required

py

int | None

The padding value to be added to both the top and bottom sides of each bounding box. If not provided, px will be used for both dimensions.

 None

Returns:

Type Description
NDArray[number]

A numpy array of shape (N, 4) where each row corresponds to a bounding box with coordinates padded according to the provided padding values. Integer inputs produce int64 output, or float64 if a padded coordinate exceeds the int64 range. Floating-point inputs preserve their dtype.

Examples:

>>> import numpy as np
>>> import supervision as sv
>>> xyxy = np.array([
...     [10, 20, 30, 40],
...     [15, 25, 35, 45]
... ])
>>> sv.pad_boxes(xyxy=xyxy, px=5, py=10)
array([[ 5, 10, 35, 50],
       [10, 15, 40, 55]])
Source code in src/supervision/detection/utils/boxes.py 
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
def pad_boxes(
    xyxy: npt.NDArray[np.number],
    px: int,
    py: int | None = None,
) -> npt.NDArray[np.number]:
    """Pads bounding boxes coordinates with a constant padding.

    Args:
        xyxy: A numpy array of shape `(N, 4)` where each
            row corresponds to a bounding box in the format
            `(x_min, y_min, x_max, y_max)`.
        px: The padding value to be added to both the left and right sides of
            each bounding box.
        py: The padding value to be added to both the top and bottom
            sides of each bounding box. If not provided, `px` will be used for both
            dimensions.

    Returns:
        A numpy array of shape `(N, 4)` where each row corresponds to a
            bounding box with coordinates padded according to the provided padding
            values. Integer inputs produce `int64` output, or `float64` if a padded
            coordinate exceeds the `int64` range. Floating-point inputs preserve
            their dtype.

    Examples:
        ```pycon
        >>> import numpy as np
        >>> import supervision as sv
        >>> xyxy = np.array([
        ...     [10, 20, 30, 40],
        ...     [15, 25, 35, 45]
        ... ])
        >>> sv.pad_boxes(xyxy=xyxy, px=5, py=10)
        array([[ 5, 10, 35, 50],
               [10, 15, 40, 55]])

        ```
    """
    if py is None:
        py = px

    integer_input = np.issubdtype(xyxy.dtype, np.integer)
    padding: list[int] | npt.NDArray[Any] = [px, py]
    if integer_input:
        px, py = int(px), int(py)
        exact_integer_limit = np.iinfo(np.int64).max // 2
        is_safe_to_cast = max(abs(px), abs(py)) <= exact_integer_limit and np.all(
            xyxy <= exact_integer_limit
            if np.issubdtype(xyxy.dtype, np.unsignedinteger)
            else (xyxy >= -exact_integer_limit) & (xyxy <= exact_integer_limit)
        )
        # Use exact Python integers only when int64 intermediates could overflow.
        result = xyxy.astype(np.int64 if is_safe_to_cast else object)
        padding = np.asarray([px, py], dtype=result.dtype)
    else:
        result = cast(npt.NDArray[Any], xyxy.copy())
    result[:, [0, 1]] -= padding
    result[:, [2, 3]] += padding

    if integer_input and result.dtype == object:
        limits = np.iinfo(np.int64)
        fits_int64 = np.all((result >= limits.min) & (result <= limits.max))
        return cast(
            npt.NDArray[np.number],
            np.asarray(result, dtype=np.int64 if fits_int64 else np.float64),
        )
    return cast(npt.NDArray[np.number], result)

denormalize_boxes

supervision.detection.utils.boxes.denormalize_boxes(xyxy: npt.NDArray[np.number], resolution_wh: tuple[int, int], normalization_factor: float = 1.0) -> npt.NDArray[np.number]

Convert normalized bounding box coordinates to absolute pixel coordinates.

Multiplies each bounding box coordinate by image size and divides by normalization_factor, mapping values from normalized [0, normalization_factor] to absolute pixel values for a given resolution.

Parameters:

Name Type Description Default

xyxy

NDArray[number]

Normalized bounding boxes of shape (N, 4), where each row is (x_min, y_min, x_max, y_max), values in [0, normalization_factor].

 required

resolution_wh

tuple[int, int]

Target image resolution as (width, height).

 required

normalization_factor

float

Maximum value of input coordinate range. Defaults to 1.0.

 1.0

Returns:

Type Description
NDArray[number]

Array of shape (N, 4) with absolute coordinates in (x_min, y_min, x_max, y_max) format.

Examples:

>>> import numpy as np
>>> import supervision as sv
>>> xyxy = np.array([
...     [0.1, 0.2, 0.5, 0.6],
...     [0.3, 0.4, 0.7, 0.8],
...     [0.2, 0.1, 0.6, 0.5]
... ])
>>> sv.denormalize_boxes(xyxy, (1280, 720))
array([[128., 144., 640., 432.],
       [384., 288., 896., 576.],
       [256.,  72., 768., 360.]])

>>> xyxy = np.array([
...     [256., 128., 768., 640.]
... ])
>>> sv.denormalize_boxes(xyxy, (1280, 720), normalization_factor=1024.0)
array([[320.,  90., 960., 450.]])
Source code in src/supervision/detection/utils/boxes.py 
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
def denormalize_boxes(
    xyxy: npt.NDArray[np.number],
    resolution_wh: tuple[int, int],
    normalization_factor: float = 1.0,
) -> npt.NDArray[np.number]:
    """Convert normalized bounding box coordinates to absolute pixel coordinates.

    Multiplies each bounding box coordinate by image size and divides by
    `normalization_factor`, mapping values from normalized `[0, normalization_factor]`
    to absolute pixel values for a given resolution.

    Args:
        xyxy: Normalized bounding boxes of shape `(N, 4)`,
            where each row is `(x_min, y_min, x_max, y_max)`, values in
            `[0, normalization_factor]`.
        resolution_wh: Target image resolution as `(width, height)`.
        normalization_factor: Maximum value of input coordinate range.
            Defaults to `1.0`.

    Returns:
        Array of shape `(N, 4)` with absolute coordinates in
            `(x_min, y_min, x_max, y_max)` format.

    Examples:
        ```pycon
        >>> import numpy as np
        >>> import supervision as sv
        >>> xyxy = np.array([
        ...     [0.1, 0.2, 0.5, 0.6],
        ...     [0.3, 0.4, 0.7, 0.8],
        ...     [0.2, 0.1, 0.6, 0.5]
        ... ])
        >>> sv.denormalize_boxes(xyxy, (1280, 720))
        array([[128., 144., 640., 432.],
               [384., 288., 896., 576.],
               [256.,  72., 768., 360.]])

        ```

        ```pycon
        >>> xyxy = np.array([
        ...     [256., 128., 768., 640.]
        ... ])
        >>> sv.denormalize_boxes(xyxy, (1280, 720), normalization_factor=1024.0)
        array([[320.,  90., 960., 450.]])

        ```
    """
    width, height = resolution_wh

    # Preserve floating caller dtypes; integer inputs need float64 to retain
    # fractional pixel coordinates rather than silently truncating them.
    scale_dtype = (
        xyxy.dtype if np.issubdtype(xyxy.dtype, np.inexact) else np.dtype(np.float64)
    )
    scale = np.array([width, height, width, height], dtype=scale_dtype)
    scale /= normalization_factor

    return xyxy * scale

xyxyxyxy_to_xyxy

supervision.detection.utils.boxes.xyxyxyxy_to_xyxy(xyxyxyxy: npt.NDArray[np.number]) -> npt.NDArray[np.number]

Convert oriented bounding box corners to axis-aligned bounding boxes.

Parameters:

Name Type Description Default

xyxyxyxy

NDArray[number]

OBB corner coordinates with shape (N, 4, 2) where each box is represented as [[x1, y1], [x2, y2], [x3, y3], [x4, y4]].

 required

Returns:

Type Description
NDArray[number]

Axis-aligned bounding boxes as an array of shape (N, 4) in (x_min, y_min, x_max, y_max) format.

Raises:

Type Description
ValueError

If xyxyxyxy does not have shape (N, 4, 2).

Examples:

>>> import numpy as np
>>> import supervision as sv
>>> corners = np.array([
...     [[0, 0], [10, 0], [10, 5], [0, 5]],
...     [[5, 5], [15, 5], [15, 10], [5, 10]],
... ], dtype=np.float32)
>>> sv.xyxyxyxy_to_xyxy(corners)
array([[ 0.,  0., 10.,  5.],
       [ 5.,  5., 15., 10.]], dtype=float32)
Source code in src/supervision/detection/utils/boxes.py 
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
def xyxyxyxy_to_xyxy(
    xyxyxyxy: npt.NDArray[np.number],
) -> npt.NDArray[np.number]:
    """Convert oriented bounding box corners to axis-aligned bounding boxes.

    Args:
        xyxyxyxy: OBB corner coordinates with shape `(N, 4, 2)` where each
            box is represented as `[[x1, y1], [x2, y2], [x3, y3], [x4, y4]]`.

    Returns:
        Axis-aligned bounding boxes as an array of shape `(N, 4)`
            in `(x_min, y_min, x_max, y_max)` format.

    Raises:
        ValueError: If `xyxyxyxy` does not have shape `(N, 4, 2)`.

    Examples:
        ```pycon
        >>> import numpy as np
        >>> import supervision as sv
        >>> corners = np.array([
        ...     [[0, 0], [10, 0], [10, 5], [0, 5]],
        ...     [[5, 5], [15, 5], [15, 10], [5, 10]],
        ... ], dtype=np.float32)
        >>> sv.xyxyxyxy_to_xyxy(corners)
        array([[ 0.,  0., 10.,  5.],
               [ 5.,  5., 15., 10.]], dtype=float32)

        ```
    """
    xyxyxyxy = cast(npt.NDArray[np.number], np.asarray(xyxyxyxy))
    if xyxyxyxy.ndim != 3 or xyxyxyxy.shape[-2:] != (4, 2):
        raise ValueError(f"xyxyxyxy must have shape (N, 4, 2); got {xyxyxyxy.shape}")
    x_min = xyxyxyxy[..., 0].min(axis=-1)
    y_min = xyxyxyxy[..., 1].min(axis=-1)
    x_max = xyxyxyxy[..., 0].max(axis=-1)
    y_max = xyxyxyxy[..., 1].max(axis=-1)
    return cast(npt.NDArray[np.number], np.stack([x_min, y_min, x_max, y_max], axis=-1))

Comments