TemplateMatching#

Functions#

NVCVStatus pvaTemplateMatchingCreate(NVCVOperatorHandle *handle, NVCVTensorRequirements const *const inImageTensorRequirements, NVCVTensorRequirements const *const inTemplateTensorRequirements, NVCVTensorRequirements const *const outTensorRequirements)

Constructs and an instance of the TemplateMatching operator.

NVCVStatus pvaTemplateMatchingSubmit(NVCVOperatorHandle handle, cudaStream_t stream, NVCVTensorHandle inImage, NVCVTensorHandle inTemplate, NVCVTensorHandle out)

Submits the TemplateMatching operator to a CUDA stream.

NVCVStatus pvaTemplateMatchingSubmit(NVCVOperatorHandle handle, cupvaStream_t stream, NVCVTensorHandle inImage, NVCVTensorHandle inTemplate, NVCVTensorHandle out)

Submits the TemplateMatching operator to a cuPVA stream.

Functions#

NVCVStatus pvaTemplateMatchingCreate(
NVCVOperatorHandle *handle,
NVCVTensorRequirements const *const inImageTensorRequirements,
NVCVTensorRequirements const *const inTemplateTensorRequirements,
NVCVTensorRequirements const *const outTensorRequirements,
)#

Constructs and an instance of the TemplateMatching operator.

Limitations:

Input image tensor:

  • If the input image tensor has a resolution W × H, then W >= 94 and H >= 94.

  • Data Layout: [CHW], [HWC], [NCHW], [NHWC] only when

    • C: [1]

    • N: [1]

Data Type

Allowed

8bit Unsigned

Yes

8bit Signed

No

16bit Unsigned

No

16bit Signed

No

32bit Unsigned

No

32bit Signed

No

32bit Float

No

64bit Float

No

Input template tensor:

  • If the template image tensor has a resolution w × h, then 0 < w <= 31 and 0 < h <= 31.

  • The template is matched only over translations, so matching is neither rotation- nor scale-invariant. A rotated or resized instance of the template is not detected.

  • Data Layout: [CHW], [HWC], [NCHW], [NHWC] only when

    • C: [1]

    • N: [1]

Data Type

Allowed

8bit Unsigned

Yes

8bit Signed

No

16bit Unsigned

No

16bit Signed

No

32bit Unsigned

No

32bit Signed

No

32bit Float

No

64bit Float

No

Output tensor:

  • The output tensor is the normalized sum of squared differences (score) for every possible location of the template inside the input image tensor. The best match is the lowest score, and an exact match scores zero.

  • If the input image tensor has a resolution W × H and input template tensor has a resolution w × h, then the output image has a resolution (W - w + 1) × (H - h + 1).

  • Data Layout: [CHW], [HWC], [NCHW], [NHWC] only when

    • C: [1]

    • N: [1]

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

No

Input/Output dependency

Property

Input == Output

Data Layout

Yes

Data Type

No

Number

Yes

Channels

Yes

Width

No

Height

No

Parameters:
  • handle – [out] Where the operator instance handle will be written to.

    • Must not be NULL.

  • inImageTensorRequirements – [in] Pointer to the NVCVTensorRequirements structure for input image tensor which contains Tensor rank, shape, layout and data type information.

  • inTemplateTensorRequirements – [in] Pointer to the NVCVTensorRequirements structure for input template tensor which contains Tensor rank, shape, layout and data type information.

  • outTensorRequirements – [in] Pointer to the NVCVTensorRequirements structure for output tensor which contains Tensor rank, shape, layout and data type information.

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 pvaTemplateMatchingSubmit(
NVCVOperatorHandle handle,
cudaStream_t stream,
NVCVTensorHandle inImage,
NVCVTensorHandle inTemplate,
NVCVTensorHandle out,
)#

Submits the TemplateMatching operator to a CUDA stream.

Note

CUDA stream support requirements:

  • PVA SDK 2.7.0 or later

  • Jetpack 7 or later

  • DriveOS 7 or later

  • x86 Emulator is not supported

Parameters:
  • handle – [in] Handle to the operator.

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

  • inImage – [in] Input image tensor handle.

  • inTemplate – [in] Input template tensor handle.

  • out – [out] Output tensor handle.

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 pvaTemplateMatchingSubmit(
NVCVOperatorHandle handle,
cupvaStream_t stream,
NVCVTensorHandle inImage,
NVCVTensorHandle inTemplate,
NVCVTensorHandle out,
)#

Submits the TemplateMatching operator to a cuPVA stream.

Parameters:
  • handle – [in] Handle to the operator.

    • Must not be NULL.

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

  • inImage – [in] Input image tensor handle.

  • inTemplate – [in] Input template tensor handle.

  • out – [out] Output tensor handle.

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.