Color Twist

group NVCV_C_ALGORITHM_COLOR_TWIST

Functions

NVCVStatus cvcudaColorTwistCreate(NVCVOperatorHandle *handle)

Constructs an instance of the ColorTwist operator.

Parameters:

handle – [out] Where the image instance handle will be written to.

  • Must not be NULL.

Return values:

  • NVCV_ERROR_INVALID_ARGUMENT – Handle is null.

  • NVCV_ERROR_OUT_OF_MEMORY – Not enough memory to create the operator.

  • NVCV_SUCCESS – Operation executed successfully.

NVCVStatus cvcudaColorTwistSubmit(NVCVOperatorHandle handle, cudaStream_t stream, NVCVTensorHandle in, NVCVTensorHandle out, NVCVTensorHandle twist)

Executes the ColorTwist operation on the given cuda stream. This operation does not wait for completion.

ColorTwist modifies colors in the image by applying custom affine transformation to the channels. For the 3x4 twist matrix and 3 channel input, the output channels are: 

  output[0] = twist[0][0] * input[0] + twist[0][1] * input[1] + twist[0][2] * input[2] + twist[0][3]
  output[1] = twist[1][0] * input[0] + twist[1][1] * input[1] + twist[1][2] * input[2] + twist[1][3]
  output[2] = twist[2][0] * input[0] + twist[2][1] * input[1] + twist[2][2] * input[2] + twist[2][3]
If the input images have 4 channels, the last channel is assumed to be alpha and is copied unmodified to the output.

Limitations:

Input: Data Layout: [NVCV_TENSOR_NHWC, NVCV_TENSOR_HWC] Channels: [3, 4]

Data Type

Allowed

8bit Unsigned

Yes

8bit Signed

No

16bit Unsigned

Yes

16bit Signed

Yes

32bit Unsigned

Yes

32bit Signed

Yes

32bit Float

Yes

64bit Float

No

Output: Data Layout: [NVCV_TENSOR_NHWC, NVCV_TENSOR_HWC] Channels: [3, 4]

Data Type

Allowed

8bit Unsigned

Yes

8bit Signed

No

16bit Unsigned

Yes

16bit Signed

Yes

32bit Unsigned

Yes

32bit Signed

Yes

32bit Float

Yes

64bit Float

No

Twist matrix: Data Layout: [NVCV_TENSOR_NHW, NVCV_TENSOR_HW]

Data Type

Allowed

8bit Unsigned

No

8bit Signed

No

16bit Unsigned

No

16bit Signed

No

32bit Unsigned

No

32bit Signed

No

32bit Float

Yes

64bit Float

Yes

Input/Output dependency

Property

Input == Output

Data Layout

Yes

Data Type

Yes

Channels

Yes

Width

Yes

Height

Yes

Samples

Yes

Input/Twist matrix type dependency

Input type

Accepted twist type

uint8

float32

int16, uint16

float32

int32, uint32

float64

float32

float32

Parameters:

  • handle – [in] Handle to the operator.

    • Must not be NULL.

  • stream – [in] Handle to a valid CUDA stream.

  • in – [in] Input tensor to get values from.

    • Must not be NULL.

  • out – [out] Output tensor to set values to.

    • Must not be NULL.

  • twist – [in] Tensor describing the 3x4 affine transformation to apply to the channels extent. It can either be a 2D 3x4 tensor describing a single transformation for all samples or 3D tensor that defines the transformations for each sample separately.

Return values:

  • NVCV_ERROR_INVALID_ARGUMENT – Some parameter is outside valid range.

  • NVCV_ERROR_INTERNAL – Internal error in the operator, invalid types passed in.

  • NVCV_SUCCESS – Operation executed successfully.

NVCVStatus cvcudaColorTwistVarShapeSubmit(NVCVOperatorHandle handle, cudaStream_t stream, NVCVImageBatchHandle in, NVCVImageBatchHandle out, NVCVTensorHandle twist)

Executes the ColorTwist operation on a batch of images.

Apart from input and output image batches, all parameters are the same as cvcudaColorTwistSubmit.

Parameters:

  • in – [in] Input image batch.

  • out – [out] Output image batch.

  • twist – [in] Tensor describing the affine transformation to apply to the channels extent.