diff --git a/torchvision/transforms/functional.py b/torchvision/transforms/functional.py
index 7b950b0c45b..ad61dd234e4 100644
--- a/torchvision/transforms/functional.py
+++ b/torchvision/transforms/functional.py
@@ -63,7 +63,7 @@ def _interpolation_modes_from_int(i: int) -> InterpolationMode:
 _is_pil_image = F_pil._is_pil_image
 
 
-def get_dimensions(img: Tensor) -> list[int]:
+def get_dimensions(img: Union[Tensor, PILImage]) -> list[int]:
     """Returns the dimensions of an image as [channels, height, width].
 
     Args:
@@ -80,7 +80,7 @@ def get_dimensions(img: Tensor) -> list[int]:
     return F_pil.get_dimensions(img)
 
 
-def get_image_size(img: Tensor) -> list[int]:
+def get_image_size(img: Union[Tensor, PILImage]) -> list[int]:
     """Returns the size of an image as [width, height].
 
     Args:
@@ -97,7 +97,7 @@ def get_image_size(img: Tensor) -> list[int]:
     return F_pil.get_image_size(img)
 
 
-def get_image_num_channels(img: Tensor) -> int:
+def get_image_num_channels(img: Union[Tensor, PILImage]) -> int:
     """Returns the number of channels of an image.
 
     Args:
@@ -178,7 +178,7 @@ def to_tensor(pic: Union[PILImage, np.ndarray]) -> Tensor:
         return img
 
 
-def pil_to_tensor(pic: Any) -> Tensor:
+def pil_to_tensor(pic: PILImage) -> Tensor:
     """Convert a ``PIL Image`` to a tensor of the same type.
     This function does not support torchscript.
 
@@ -385,12 +385,12 @@ def _compute_resized_output_size(
 
 
 def resize(
-    img: Tensor,
+    img: Union[Tensor, PILImage],
     size: list[int],
     interpolation: InterpolationMode = InterpolationMode.BILINEAR,
     max_size: Optional[int] = None,
     antialias: Optional[bool] = True,
-) -> Tensor:
+) -> Union[Tensor, PILImage]:
     r"""Resize the input image to the given size.
     If the image is torch Tensor, it is expected
     to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions
@@ -479,7 +479,7 @@ def resize(
     return F_t.resize(img, size=output_size, interpolation=interpolation.value, antialias=antialias)
 
 
-def pad(img: Tensor, padding: list[int], fill: Union[int, float] = 0, padding_mode: str = "constant") -> Tensor:
+def pad(img: Union[Tensor, PILImage], padding: list[int], fill: Union[int, float] = 0, padding_mode: str = "constant") -> Union[Tensor, PILImage]:
     r"""Pad the given image on all sides with the given "pad" value.
     If the image is torch Tensor, it is expected
     to have [..., H, W] shape, where ... means at most 2 leading dimensions for mode reflect and symmetric,
@@ -528,7 +528,7 @@ def pad(img: Tensor, padding: list[int], fill: Union[int, float] = 0, padding_mo
     return F_t.pad(img, padding=padding, fill=fill, padding_mode=padding_mode)
 
 
-def crop(img: Tensor, top: int, left: int, height: int, width: int) -> Tensor:
+def crop(img: Union[Tensor, PILImage], top: int, left: int, height: int, width: int) -> Union[Tensor, PILImage]:
     """Crop the given image at specified location and output size.
     If the image is torch Tensor, it is expected
     to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.
@@ -651,7 +651,7 @@ def resized_crop(
     return img
 
 
-def hflip(img: Tensor) -> Tensor:
+def hflip(img: Union[Tensor, PILImage]) -> Union[Tensor, PILImage]:
     """Horizontally flip the given image.
 
     Args:
@@ -705,12 +705,12 @@ def _get_perspective_coeffs(startpoints: list[list[int]], endpoints: list[list[i
 
 
 def perspective(
-    img: Tensor,
+    img: Union[Tensor, PILImage],
     startpoints: list[list[int]],
     endpoints: list[list[int]],
     interpolation: InterpolationMode = InterpolationMode.BILINEAR,
     fill: Optional[list[float]] = None,
-) -> Tensor:
+) -> Union[Tensor, PILImage]:
     """Perform perspective transform of the given image.
     If the image is torch Tensor, it is expected
     to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.
@@ -754,7 +754,7 @@ def perspective(
     return F_t.perspective(img, coeffs, interpolation=interpolation.value, fill=fill)
 
 
-def vflip(img: Tensor) -> Tensor:
+def vflip(img: Union[Tensor, PILImage]) -> Union[Tensor, PILImage]:
     """Vertically flip the given image.
 
     Args:
@@ -865,7 +865,7 @@ def ten_crop(
     return first_five + second_five
 
 
-def adjust_brightness(img: Tensor, brightness_factor: float) -> Tensor:
+def adjust_brightness(img: Union[Tensor, PILImage], brightness_factor: float) -> Union[Tensor, PILImage]:
     """Adjust brightness of an image.
 
     Args:
@@ -887,7 +887,7 @@ def adjust_brightness(img: Tensor, brightness_factor: float) -> Tensor:
     return F_t.adjust_brightness(img, brightness_factor)
 
 
-def adjust_contrast(img: Tensor, contrast_factor: float) -> Tensor:
+def adjust_contrast(img: Union[Tensor, PILImage], contrast_factor: float) -> Union[Tensor, PILImage]:
     """Adjust contrast of an image.
 
     Args:
@@ -909,7 +909,7 @@ def adjust_contrast(img: Tensor, contrast_factor: float) -> Tensor:
     return F_t.adjust_contrast(img, contrast_factor)
 
 
-def adjust_saturation(img: Tensor, saturation_factor: float) -> Tensor:
+def adjust_saturation(img: Union[Tensor, PILImage], saturation_factor: float) -> Union[Tensor, PILImage]:
     """Adjust color saturation of an image.
 
     Args:
@@ -931,7 +931,7 @@ def adjust_saturation(img: Tensor, saturation_factor: float) -> Tensor:
     return F_t.adjust_saturation(img, saturation_factor)
 
 
-def adjust_hue(img: Tensor, hue_factor: float) -> Tensor:
+def adjust_hue(img: Union[Tensor, PILImage], hue_factor: float) -> Union[Tensor, PILImage]:
     """Adjust hue of an image.
 
     The image hue is adjusted by converting the image to HSV and
@@ -970,7 +970,7 @@ def adjust_hue(img: Tensor, hue_factor: float) -> Tensor:
     return F_t.adjust_hue(img, hue_factor)
 
 
-def adjust_gamma(img: Tensor, gamma: float, gain: float = 1) -> Tensor:
+def adjust_gamma(img: Union[Tensor, PILImage], gamma: float, gain: float = 1) -> Union[Tensor, PILImage]:
     r"""Perform gamma correction on an image.
 
     Also known as Power Law Transform. Intensities in RGB mode are adjusted
@@ -1064,13 +1064,13 @@ def _get_inverse_affine_matrix(
 
 
 def rotate(
-    img: Tensor,
+    img: Union[Tensor, PILImage],
     angle: float,
     interpolation: InterpolationMode = InterpolationMode.NEAREST,
     expand: bool = False,
     center: Optional[list[int]] = None,
     fill: Optional[list[float]] = None,
-) -> Tensor:
+) -> Union[Tensor, PILImage]:
     """Rotate the image by angle.
     If the image is torch Tensor, it is expected
     to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.
@@ -1133,7 +1133,7 @@ def rotate(
 
 
 def affine(
-    img: Tensor,
+    img: Union[Tensor, PILImage],
     angle: float,
     translate: list[int],
     scale: float,
@@ -1141,7 +1141,7 @@ def affine(
     interpolation: InterpolationMode = InterpolationMode.NEAREST,
     fill: Optional[list[float]] = None,
     center: Optional[list[int]] = None,
-) -> Tensor:
+) -> Union[Tensor, PILImage]:
     """Apply affine transformation on the image keeping image center invariant.
     If the image is torch Tensor, it is expected
     to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.
@@ -1264,7 +1264,7 @@ def to_grayscale(img, num_output_channels=1):
     raise TypeError("Input should be PIL Image")
 
 
-def rgb_to_grayscale(img: Tensor, num_output_channels: int = 1) -> Tensor:
+def rgb_to_grayscale(img: Union[Tensor, PILImage], num_output_channels: int = 1) -> Union[Tensor, PILImage]:
     """Convert RGB image to grayscale version of image.
     If the image is torch Tensor, it is expected
     to have [..., 3, H, W] shape, where ... means an arbitrary number of leading dimensions
@@ -1315,7 +1315,7 @@ def erase(img: Tensor, i: int, j: int, h: int, w: int, v: Tensor, inplace: bool
     return F_t.erase(img, i, j, h, w, v, inplace=inplace)
 
 
-def gaussian_blur(img: Tensor, kernel_size: list[int], sigma: Optional[list[float]] = None) -> Tensor:
+def gaussian_blur(img: Union[Tensor, PILImage], kernel_size: list[int], sigma: Optional[list[float]] = None) -> Union[Tensor, PILImage]:
     """Performs Gaussian blurring on the image by given kernel
 
     The convolution will be using reflection padding corresponding to the kernel size, to maintain the input shape.
@@ -1384,7 +1384,7 @@ def gaussian_blur(img: Tensor, kernel_size: list[int], sigma: Optional[list[floa
     return output
 
 
-def invert(img: Tensor) -> Tensor:
+def invert(img: Union[Tensor, PILImage]) -> Union[Tensor, PILImage]:
     """Invert the colors of an RGB/grayscale image.
 
     Args:
@@ -1404,7 +1404,7 @@ def invert(img: Tensor) -> Tensor:
     return F_t.invert(img)
 
 
-def posterize(img: Tensor, bits: int) -> Tensor:
+def posterize(img: Union[Tensor, PILImage], bits: int) -> Union[Tensor, PILImage]:
     """Posterize an image by reducing the number of bits for each color channel.
 
     Args:
@@ -1428,7 +1428,7 @@ def posterize(img: Tensor, bits: int) -> Tensor:
     return F_t.posterize(img, bits)
 
 
-def solarize(img: Tensor, threshold: float) -> Tensor:
+def solarize(img: Union[Tensor, PILImage], threshold: float) -> Union[Tensor, PILImage]:
     """Solarize an RGB/grayscale image by inverting all pixel values above a threshold.
 
     Args:
@@ -1448,7 +1448,7 @@ def solarize(img: Tensor, threshold: float) -> Tensor:
     return F_t.solarize(img, threshold)
 
 
-def adjust_sharpness(img: Tensor, sharpness_factor: float) -> Tensor:
+def adjust_sharpness(img: Union[Tensor, PILImage], sharpness_factor: float) -> Union[Tensor, PILImage]:
     """Adjust the sharpness of an image.
 
     Args:
@@ -1470,7 +1470,7 @@ def adjust_sharpness(img: Tensor, sharpness_factor: float) -> Tensor:
     return F_t.adjust_sharpness(img, sharpness_factor)
 
 
-def autocontrast(img: Tensor) -> Tensor:
+def autocontrast(img: Union[Tensor, PILImage]) -> Union[Tensor, PILImage]:
     """Maximize contrast of an image by remapping its
     pixels per channel so that the lowest becomes black and the lightest
     becomes white.
@@ -1492,7 +1492,7 @@ def autocontrast(img: Tensor) -> Tensor:
     return F_t.autocontrast(img)
 
 
-def equalize(img: Tensor) -> Tensor:
+def equalize(img: Union[Tensor, PILImage]) -> Union[Tensor, PILImage]:
     """Equalize the histogram of an image by applying
     a non-linear mapping to the input in order to create a uniform
     distribution of grayscale values in the output.
@@ -1516,7 +1516,7 @@ def equalize(img: Tensor) -> Tensor:
 
 
 def elastic_transform(
-    img: Tensor,
+    img: Union[Tensor, PILImage],
     displacement: Tensor,
     interpolation: InterpolationMode = InterpolationMode.BILINEAR,
     fill: Optional[list[float]] = None,