ImageProjectiveTransformV3

public final class ImageProjectiveTransformV3

Applies the given transform to each of the images.

If one row of `transforms` is `[a0, a1, a2, b0, b1, b2, c0, c1]`, then it maps the output point `(x, y)` to a transformed input point `(x', y') = ((a0 x + a1 y + a2) / k, (b0 x + b1 y + b2) / k)`, where `k = c0 x + c1 y + 1`. If the transformed point lays outside of the input image, the output pixel is set to fill_value.

Nested Classes

class ImageProjectiveTransformV3.Options Optional attributes for ImageProjectiveTransformV3

Constants

String OP_NAME The name of this op, as known by TensorFlow core engine

Public Methods

Output <T>
asOutput ()
Returns the symbolic handle of the tensor.
static <T extends TNumber > ImageProjectiveTransformV3 <T>
create ( Scope scope, Operand <T> images, Operand < TFloat32 > transforms, Operand < TInt32 > outputShape, Operand < TFloat32 > fillValue, String interpolation, Options... options)
Factory method to create a class wrapping a new ImageProjectiveTransformV3 operation.
static ImageProjectiveTransformV3.Options
fillMode (String fillMode)
Output <T>
transformedImages ()
4-D with shape `[batch, new_height, new_width, channels]`.

Inherited Methods

Constants

public static final String OP_NAME

The name of this op, as known by TensorFlow core engine

Constant Value: "ImageProjectiveTransformV3"

Public Methods

public Output <T> asOutput ()

Returns the symbolic handle of the tensor.

Inputs to TensorFlow operations are outputs of another TensorFlow operation. This method is used to obtain a symbolic handle that represents the computation of the input.

public static ImageProjectiveTransformV3 <T> create ( Scope scope, Operand <T> images, Operand < TFloat32 > transforms, Operand < TInt32 > outputShape, Operand < TFloat32 > fillValue, String interpolation, Options... options)

Factory method to create a class wrapping a new ImageProjectiveTransformV3 operation.

Parameters
scope current scope
images 4-D with shape `[batch, height, width, channels]`.
transforms 2-D Tensor, `[batch, 8]` or `[1, 8]` matrix, where each row corresponds to a 3 x 3 projective transformation matrix, with the last entry assumed to be 1. If there is one row, the same transformation will be applied to all images.
outputShape 1-D Tensor [new_height, new_width].
fillValue float, the value to be filled when fill_mode is constant".
interpolation Interpolation method, "NEAREST" or "BILINEAR".
options carries optional attributes values
Returns
  • a new instance of ImageProjectiveTransformV3

public static ImageProjectiveTransformV3.Options fillMode (String fillMode)

Parameters
fillMode Fill mode, "REFLECT", "WRAP", or "CONSTANT".

public Output <T> transformedImages ()

4-D with shape `[batch, new_height, new_width, channels]`.