RPPT Tensor Operations - Geometric Augmentations.#

RPP: RPPT Tensor Operations - Geometric Augmentations.
RPPT Tensor Operations - Geometric Augmentations.

RPPT Tensor Operations - Geometric Augmentations. More...

Functions

RppStatus rppt_crop_host (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Crop augmentation on HOST backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_crop_gpu (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Crop augmentation on HIP backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_crop_mirror_normalize_host (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32f *offsetTensor, Rpp32f *multiplierTensor, Rpp32u *mirrorTensor, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Crop mirror normalize augmentation on HOST backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_crop_mirror_normalize_gpu (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32f *offsetTensor, Rpp32f *multiplierTensor, Rpp32u *mirrorTensor, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Crop mirror normalize augmentation on HIP backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_warp_affine_host (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32f *affineTensor, RpptInterpolationType interpolationType, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Warp affine augmentation on HOST backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_warp_affine_gpu (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32f *affineTensor, RpptInterpolationType interpolationType, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Warp affine augmentation on HIP backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_flip_host (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32u *horizontalTensor, Rpp32u *verticalTensor, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Flip augmentation on HOST backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_flip_gpu (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32u *horizontalTensor, Rpp32u *verticalTensor, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Flip augmentation on HIP backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_resize_host (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptImagePatchPtr dstImgSizes, RpptInterpolationType interpolationType, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Resize augmentation on HOST backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_resize_gpu (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptImagePatchPtr dstImgSizes, RpptInterpolationType interpolationType, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Resize augmentation on HIP backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_resize_mirror_normalize_host (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptImagePatchPtr dstImgSizes, RpptInterpolationType interpolationType, Rpp32f *meanTensor, Rpp32f *stdDevTensor, Rpp32u *mirrorTensor, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Resize mirror normalize augmentation on HOST backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_resize_mirror_normalize_gpu (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptImagePatchPtr dstImgSizes, RpptInterpolationType interpolationType, Rpp32f *meanTensor, Rpp32f *stdDevTensor, Rpp32u *mirrorTensor, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Resize mirror normalize augmentation on HIP backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_resize_crop_mirror_host (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptImagePatchPtr dstImgSizes, RpptInterpolationType interpolationType, Rpp32u *mirrorTensor, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Resize crop mirror augmentation on HOST backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_resize_crop_mirror_gpu (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptImagePatchPtr dstImgSizes, RpptInterpolationType interpolationType, Rpp32u *mirrorTensor, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Resize crop mirror augmentation on HIP backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_rotate_host (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32f *angle, RpptInterpolationType interpolationType, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Rotate augmentation on HOST backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_rotate_gpu (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32f *angle, RpptInterpolationType interpolationType, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Rotate augmentation on HIP backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_phase_host (RppPtr_t srcPtr1, RppPtr_t srcPtr2, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Phase augmentation on HOST backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_phase_gpu (RppPtr_t srcPtr1, RppPtr_t srcPtr2, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Phase augmentation on HIP backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_slice_host (RppPtr_t srcPtr, RpptGenericDescPtr srcGenericDescPtr, RppPtr_t dstPtr, RpptGenericDescPtr dstGenericDescPtr, Rpp32s *anchorTensor, Rpp32s *shapeTensor, RppPtr_t fillValue, bool enablePadding, Rpp32u *roiTensor, rppHandle_t rppHandle)
 Slice augmentation HOST. More...
 
RppStatus rppt_slice_gpu (RppPtr_t srcPtr, RpptGenericDescPtr srcGenericDescPtr, RppPtr_t dstPtr, RpptGenericDescPtr dstGenericDescPtr, Rpp32s *anchorTensor, Rpp32s *shapeTensor, RppPtr_t fillValue, bool enablePadding, Rpp32u *roiTensor, rppHandle_t rppHandle)
 Slice augmentation GPU. More...
 
RppStatus rppt_crop_and_patch_host (RppPtr_t srcPtr1, RppPtr_t srcPtr2, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptROIPtr roiTensorPtrDst, RpptROIPtr cropRoi, RpptROIPtr patchRoi, RpptRoiType roiType, rppHandle_t rppHandle)
 Crop and Patch augmentation on HOST backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_crop_and_patch_gpu (RppPtr_t srcPtr1, RppPtr_t srcPtr2, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptROIPtr roiTensorPtrDst, RpptROIPtr cropRoi, RpptROIPtr patchRoi, RpptRoiType roiType, rppHandle_t rppHandle)
 Crop and Patch augmentation on HIP backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_flip_voxel_host (RppPtr_t srcPtr, RpptGenericDescPtr srcGenericDescPtr, RppPtr_t dstPtr, RpptGenericDescPtr dstGenericDescPtr, Rpp32u *horizontalTensor, Rpp32u *verticalTensor, Rpp32u *depthTensor, RpptROI3DPtr roiGenericPtrSrc, RpptRoi3DType roiType, rppHandle_t rppHandle)
 Flip voxel augmentation HOST. More...
 
RppStatus rppt_flip_voxel_gpu (RppPtr_t srcPtr, RpptGenericDescPtr srcGenericDescPtr, RppPtr_t dstPtr, RpptGenericDescPtr dstGenericDescPtr, Rpp32u *horizontalTensor, Rpp32u *verticalTensor, Rpp32u *depthTensor, RpptROI3DPtr roiGenericPtrSrc, RpptRoi3DType roiType, rppHandle_t rppHandle)
 Flip voxel augmentation GPU. More...
 
RppStatus rppt_remap_host (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32f *rowRemapTable, Rpp32f *colRemapTable, RpptDescPtr tableDescPtr, RpptInterpolationType interpolationType, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Remap augmentation on HOST backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_remap_gpu (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32f *rowRemapTable, Rpp32f *colRemapTable, RpptDescPtr tableDescPtr, RpptInterpolationType interpolationType, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Remap augmentation on HIP backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_lens_correction_host (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32f *rowRemapTable, Rpp32f *colRemapTable, RpptDescPtr tableDescPtr, Rpp32f *cameraMatrixTensor, Rpp32f *distortionCoeffsTensor, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Lens correction transformation on HOST backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_lens_correction_gpu (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32f *rowRemapTable, Rpp32f *colRemapTable, RpptDescPtr tableDescPtr, Rpp32f *cameraMatrixTensor, Rpp32f *distortionCoeffsTensor, RpptROIPtr roiTensorPtrSrc, RpptRoiType roiType, rppHandle_t rppHandle)
 Lens correction transformation on HIP backend for a NCHW/NHWC layout tensor. More...
 
RppStatus rppt_transpose_host (RppPtr_t srcPtr, RpptGenericDescPtr srcGenericDescPtr, RppPtr_t dstPtr, RpptGenericDescPtr dstGenericDescPtr, Rpp32u *permTensor, Rpp32u *roiTensor, rppHandle_t rppHandle)
 Transpose Generic augmentation on HOST backend. More...
 
RppStatus rppt_transpose_gpu (RppPtr_t srcPtr, RpptGenericDescPtr srcGenericDescPtr, RppPtr_t dstPtr, RpptGenericDescPtr dstGenericDescPtr, Rpp32u *permTensor, Rpp32u *roiTensor, rppHandle_t rppHandle)
 Transpose Generic augmentation on HIP backend. More...
 

Detailed Description

RPPT Tensor Operations - Geometric Augmentations.

Function Documentation

◆ rppt_crop_and_patch_gpu()

RppStatus rppt_crop_and_patch_gpu ( RppPtr_t  srcPtr1,
RppPtr_t  srcPtr2,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
RpptROIPtr  roiTensorPtrDst,
RpptROIPtr  cropRoi,
RpptROIPtr  patchRoi,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Crop and Patch augmentation on HIP backend for a NCHW/NHWC layout tensor.

The crop and patch augmentation crops a ROI from 1st image and patches the cropped region in 2nd image as per the patch co-ordinates for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr1 depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • srcPtr2 depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input1
    Sample Input2
    Sample Output
    Parameters
    [in]srcPtr1source tensor1 in HIP memory
    [in]srcPtr2source tensor2 in HIP memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HIP memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]roiTensorPtrSrcROI data in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]cropRoiTensorcrop co-ordinates in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]patchRoiTensorpatch co-ordinates in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_crop_and_patch_host()

RppStatus rppt_crop_and_patch_host ( RppPtr_t  srcPtr1,
RppPtr_t  srcPtr2,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
RpptROIPtr  roiTensorPtrDst,
RpptROIPtr  cropRoi,
RpptROIPtr  patchRoi,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Crop and Patch augmentation on HOST backend for a NCHW/NHWC layout tensor.

The crop and patch augmentation crops a ROI from 1st image and patches the cropped region in 2nd image as per the patch co-ordinates for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr1 depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • srcPtr2 depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input1
    Sample Input2
    Sample Output
    Parameters
    [in]srcPtr1source tensor1 in HOST memory
    [in]srcPtr2source tensor2 in HOST memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HOST memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]roiTensorPtrSrcROI data in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]cropRoiTensorcrop co-ordinates in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]patchRoiTensorpatch co-ordinates in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_crop_gpu()

RppStatus rppt_crop_gpu ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Crop augmentation on HIP backend for a NCHW/NHWC layout tensor.

The crop augmentation crops each image to a given ROI, for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HIP memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HIP memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]roiTensorPtrSrcROI data in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_crop_host()

RppStatus rppt_crop_host ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Crop augmentation on HOST backend for a NCHW/NHWC layout tensor.

The crop augmentation crops each image to a given ROI, for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HOST memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HOST memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]roiTensorPtrSrcROI data in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_crop_mirror_normalize_gpu()

RppStatus rppt_crop_mirror_normalize_gpu ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
Rpp32f offsetTensor,
Rpp32f multiplierTensor,
Rpp32u mirrorTensor,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Crop mirror normalize augmentation on HIP backend for a NCHW/NHWC layout tensor.

The crop mirror normalize augmentation crops each image to a given ROI, does an optional mirror and/or normalize for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HIP memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HIP memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]offsetTensoroffset values for normalization (1D tensor in pinned/HOST memory, of size batchSize, with offsetTensor[n] <= 0)
    [in]multiplierTensormultiplier values for normalization (1D tensor in pinned/HOST memory, of size batchSize, with multiplierTensor[n] > 0)
    [in]mirrorTensormirror flag values to set mirroring on/off (1D tensor in pinned/HOST memory, of size batchSize, with mirrorTensor[n] = 0/1)
    [in]roiTensorPtrSrcROI data in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_crop_mirror_normalize_host()

RppStatus rppt_crop_mirror_normalize_host ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
Rpp32f offsetTensor,
Rpp32f multiplierTensor,
Rpp32u mirrorTensor,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Crop mirror normalize augmentation on HOST backend for a NCHW/NHWC layout tensor.

The crop mirror normalize augmentation crops each image to a given ROI, does an optional mirror and/or normalize for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HOST memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HOST memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]offsetTensoroffset values for normalization (1D tensor in HOST memory, of size batchSize, with offsetTensor[n] <= 0)
    [in]multiplierTensormultiplier values for normalization (1D tensor in HOST memory, of size batchSize, with multiplierTensor[n] > 0)
    [in]mirrorTensormirror flag values to set mirroring on/off (1D tensor in HOST memory, of size batchSize, with mirrorTensor[n] = 0/1)
    [in]roiTensorPtrSrcROI data in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_flip_gpu()

RppStatus rppt_flip_gpu ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
Rpp32u horizontalTensor,
Rpp32u verticalTensor,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Flip augmentation on HIP backend for a NCHW/NHWC layout tensor.

The flip augmentation performs a mask-controlled horizontal/vertical flip for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HIP memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HIP memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]horizontalTensorhorizontal flag values to set horizontal flip on/off (1D tensor in pinned/HOST memory, of size batchSize, with horizontalTensor[i] = 0/1)
    [in]verticalTensorvertical flag values to set vertical flip on/off (1D tensor in pinned/HOST memory, of size batchSize, with verticalTensor[i] = 0/1)
    [in]roiTensorPtrSrcROI data in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_flip_host()

RppStatus rppt_flip_host ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
Rpp32u horizontalTensor,
Rpp32u verticalTensor,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Flip augmentation on HOST backend for a NCHW/NHWC layout tensor.

The flip augmentation performs a mask-controlled horizontal/vertical flip for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HOST memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HOST memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]horizontalTensorhorizontal flag values to set horizontal flip on/off (1D tensor in HOST memory, of size batchSize, with horizontalTensor[i] = 0/1)
    [in]verticalTensorvertical flag values to set vertical flip on/off (1D tensor in HOST memory, of size batchSize, with verticalTensor[i] = 0/1)
    [in]roiTensorPtrSrcROI data in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_flip_voxel_gpu()

RppStatus rppt_flip_voxel_gpu ( RppPtr_t  srcPtr,
RpptGenericDescPtr  srcGenericDescPtr,
RppPtr_t  dstPtr,
RpptGenericDescPtr  dstGenericDescPtr,
Rpp32u horizontalTensor,
Rpp32u verticalTensor,
Rpp32u depthTensor,
RpptROI3DPtr  roiGenericPtrSrc,
RpptRoi3DType  roiType,
rppHandle_t  rppHandle 
)

Flip voxel augmentation GPU.

The flip voxel augmentation performs a mask-controlled horizontal/vertical/depth flip on a generic 4D tensor.
Support added for f32 -> f32 and u8 -> u8 dataypes.

Sample Input
Sample Output
Parameters
[in]srcPtrsource tensor in HIP memory
[in]srcGenericDescPtrsource tensor descriptor (Restrictions - numDims = 5, offsetInBytes >= 0, dataType = U8/F32, layout = NCDHW/NDHWC, c = 1/3)
[out]dstPtrdestination tensor in HIP memory
[in]dstGenericDescPtrdestination tensor descriptor (Restrictions - numDims = 5, offsetInBytes >= 0, dataType = U8/F32, layout = NCDHW/NDHWC, c = 1/3)
[in]horizontalTensorhorizontal flag values to set horizontal flip on/off (1D tensor in pinned/HOST memory, of size batchSize, with horizontalTensor[i] = 0/1)
[in]verticalTensorvertical flag values to set vertical flip on/off (1D tensor in pinned/HOST memory, of size batchSize, with verticalTensor[i] = 0/1)
[in]depthTensordepth flag values to set depth flip on/off (1D tensor in pinned/HOST memory, of size batchSize, with depthTensor[i] = 0/1)
[in]roiGenericPtrSrcROI data for each image in source tensor (tensor of batchSize RpptRoiGeneric values)
[in]roiTypeROI type used (RpptRoi3DType::XYZWHD or RpptRoi3DType::LTFRBB)
[in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
Returns
A RppStatus enumeration.
Return values
RPP_SUCCESSSuccessful completion.
RPP_ERROR*Unsuccessful completion.

◆ rppt_flip_voxel_host()

RppStatus rppt_flip_voxel_host ( RppPtr_t  srcPtr,
RpptGenericDescPtr  srcGenericDescPtr,
RppPtr_t  dstPtr,
RpptGenericDescPtr  dstGenericDescPtr,
Rpp32u horizontalTensor,
Rpp32u verticalTensor,
Rpp32u depthTensor,
RpptROI3DPtr  roiGenericPtrSrc,
RpptRoi3DType  roiType,
rppHandle_t  rppHandle 
)

Flip voxel augmentation HOST.

The flip voxel augmentation performs a mask-controlled horizontal/vertical/depth flip on a generic 4D tensor.
Support added for f32 -> f32 and u8 -> u8 dataypes.

Sample Input
Sample Output
Parameters
[in]srcPtrsource tensor in HOST memory
[in]srcGenericDescPtrsource tensor descriptor (Restrictions - numDims = 5, offsetInBytes >= 0, dataType = U8/F32, layout = NCDHW/NDHWC, c = 1/3)
[out]dstPtrdestination tensor in HOST memory
[in]dstGenericDescPtrdestination tensor descriptor (Restrictions - numDims = 5, offsetInBytes >= 0, dataType = U8/F32, layout = NCDHW/NDHWC, c = 1/3)
[in]horizontalTensorhorizontal flag values to set horizontal flip on/off (1D tensor in HOST memory, of size batchSize, with horizontalTensor[i] = 0/1)
[in]verticalTensorvertical flag values to set vertical flip on/off (1D tensor in HOST memory, of size batchSize, with verticalTensor[i] = 0/1)
[in]depthTensordepth flag values to set depth flip on/off (1D tensor in HOST memory, of size batchSize, with depthTensor[i] = 0/1)
[in]roiGenericPtrSrcROI data for each image in source tensor (tensor of batchSize RpptRoiGeneric values)
[in]roiTypeROI type used (RpptRoi3DType::XYZWHD or RpptRoi3DType::LTFRBB)
[in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
Returns
A RppStatus enumeration.
Return values
RPP_SUCCESSSuccessful completion.
RPP_ERROR*Unsuccessful completion.

◆ rppt_lens_correction_gpu()

RppStatus rppt_lens_correction_gpu ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
Rpp32f rowRemapTable,
Rpp32f colRemapTable,
RpptDescPtr  tableDescPtr,
Rpp32f cameraMatrixTensor,
Rpp32f distortionCoeffsTensor,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Lens correction transformation on HIP backend for a NCHW/NHWC layout tensor.

Performs lens correction transforms on an image to compensate barrel lens distortion of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr. Note: Returns a black image if the passed camera matrix has a 0 determinant
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HIP memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HIP memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]rowRemapTableRpp32f row numbers in HIP memory for every pixel in the input batch of images (1D tensor of size width * height * batchSize)
    [in]colRemapTableRpp32f column numbers in HIP memory for every pixel in the input batch of images (1D tensor of size width * height * batchSize)
    [in]tableDescPtrtable tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = F32, layout = NHWC, c = 1)
    [in]cameraMatrixTensorcontains camera intrinsic parameters required to compute lens corrected image. (1D tensor of size 9 * batchSize)
    [in]distortionCoeffsTensorcontains distortion coefficients required to compute lens corrected image. (1D tensor of size 8 * batchSize)
    [in]roiTensorPtrSrcROI data in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_lens_correction_host()

RppStatus rppt_lens_correction_host ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
Rpp32f rowRemapTable,
Rpp32f colRemapTable,
RpptDescPtr  tableDescPtr,
Rpp32f cameraMatrixTensor,
Rpp32f distortionCoeffsTensor,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Lens correction transformation on HOST backend for a NCHW/NHWC layout tensor.

Performs lens correction transforms on an image to compensate barrel lens distortion of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr. Note: Returns a black image if the passed camera matrix has a 0 determinant
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HOST memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HOST memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]rowRemapTableRpp32f row numbers in HOST memory for every pixel in the input batch of images (1D tensor of size width * height * batchSize)
    [in]colRemapTableRpp32f column numbers in HOST memory for every pixel in the input batch of images (1D tensor of size width * height * batchSize)
    [in]tableDescPtrtable tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = F32, layout = NHWC, c = 1)
    [in]cameraMatrixTensorcontains camera intrinsic parameters required to compute lens corrected image. (1D tensor of size 9 * batchSize)
    [in]distortionCoeffsTensorcontains distortion coefficients required to compute lens corrected image. (1D tensor of size 8 * batchSize)
    [in]roiTensorPtrSrcROI data in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_phase_gpu()

RppStatus rppt_phase_gpu ( RppPtr_t  srcPtr1,
RppPtr_t  srcPtr2,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Phase augmentation on HIP backend for a NCHW/NHWC layout tensor.

The phase augmentation computes phase of corresponding pixels for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input1
    Sample Input2
    Sample Output
    Parameters
    [in]srcPtr1source1 tensor in HIP memory
    [in]srcPtr2source2 tensor in HIP memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HIP memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]roiTensorPtrSrcROI data in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_phase_host()

RppStatus rppt_phase_host ( RppPtr_t  srcPtr1,
RppPtr_t  srcPtr2,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Phase augmentation on HOST backend for a NCHW/NHWC layout tensor.

The phase augmentation computes phase of corresponding pixels for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input1
    Sample Input2
    Sample Output
    Parameters
    [in]srcPtr1source1 tensor in HOST memory
    [in]srcPtr2source2 tensor in HOST memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HOST memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]roiTensorPtrSrcROI data in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_remap_gpu()

RppStatus rppt_remap_gpu ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
Rpp32f rowRemapTable,
Rpp32f colRemapTable,
RpptDescPtr  tableDescPtr,
RpptInterpolationType  interpolationType,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Remap augmentation on HIP backend for a NCHW/NHWC layout tensor.

Performs a remap operation using user specified remap tables for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout. For each image, the output(x,y) = input(mapx(x, y), mapy(x, y)) for every (x,y) in the destination image.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HIP memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HIP memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]rowRemapTableRpp32f row numbers in HIP memory for every pixel in the input batch of images (Restrictions - rois in the rowRemapTable data for each image in batch must match roiTensorSrc)
    [in]colRemapTableRpp32f column numbers in HIP memory for every pixel in the input batch of images (Restrictions - rois in the colRemapTable data for each image in batch must match roiTensorSrc)
    [in]tableDescPtrrowRemapTable and colRemapTable common tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = F32, layout = NHWC, c = 1)
    [in]interpolationTypeInterpolation type used in RpptInterpolationType (Restrictions - Supports only NEAREST_NEIGHBOR and BILINEAR)
    [in]roiTensorPtrSrcROI data in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_remap_host()

RppStatus rppt_remap_host ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
Rpp32f rowRemapTable,
Rpp32f colRemapTable,
RpptDescPtr  tableDescPtr,
RpptInterpolationType  interpolationType,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Remap augmentation on HOST backend for a NCHW/NHWC layout tensor.

Performs a remap operation using user specified remap tables for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout. For each image, the output(x,y) = input(mapx(x, y), mapy(x, y)) for every (x,y) in the destination image.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HOST memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HOST memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]rowRemapTableRpp32f row numbers in HOST memory for every pixel in the input batch of images (Restrictions - rois in the rowRemapTable data for each image in batch must match roiTensorSrc)
    [in]colRemapTableRpp32f column numbers in HOST memory for every pixel in the input batch of images (Restrictions - rois in the colRemapTable data for each image in batch must match roiTensorSrc)
    [in]tableDescPtrrowRemapTable and colRemapTable common tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = F32, layout = NHWC, c = 1)
    [in]interpolationTypeInterpolation type used in RpptInterpolationType (Restrictions - Supports only NEAREST_NEIGHBOR and BILINEAR)
    [in]roiTensorPtrSrcROI data in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_resize_crop_mirror_gpu()

RppStatus rppt_resize_crop_mirror_gpu ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
RpptImagePatchPtr  dstImgSizes,
RpptInterpolationType  interpolationType,
Rpp32u mirrorTensor,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Resize crop mirror augmentation on HIP backend for a NCHW/NHWC layout tensor.

The resize crop mirror augmentation performs an image resized crop, with an optional mirror, for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HIP memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HIP memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]dstImgSizesdestination image sizes ( RpptImagePatchPtr type pointer to array, in pinned/HOST memory, of size batchSize)
    [in]interpolationTypeInterpolation type used in RpptInterpolationType
    [in]mirrorTensormirror flag value to set mirroring on/off (1D tensor in pinned/HOST memory, of size batchSize, with mirrorTensor[n] = 0/1)
    [in]roiTensorPtrSrcROI data in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_resize_crop_mirror_host()

RppStatus rppt_resize_crop_mirror_host ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
RpptImagePatchPtr  dstImgSizes,
RpptInterpolationType  interpolationType,
Rpp32u mirrorTensor,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Resize crop mirror augmentation on HOST backend for a NCHW/NHWC layout tensor.

The resize crop mirror augmentation performs an image resized crop, with an optional mirror, for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HOST memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HOST memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]dstImgSizesdestination image sizes ( RpptImagePatchPtr type pointer to array, in HOST memory, of size batchSize)
    [in]interpolationTypeInterpolation type used in RpptInterpolationType
    [in]mirrorTensormirror flag value to set mirroring on/off (1D tensor in HOST memory, of size batchSize, with mirrorTensor[n] = 0/1)
    [in]roiTensorPtrSrcROI data in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_resize_gpu()

RppStatus rppt_resize_gpu ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
RpptImagePatchPtr  dstImgSizes,
RpptInterpolationType  interpolationType,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Resize augmentation on HIP backend for a NCHW/NHWC layout tensor.

The resize augmentation performs an image resize for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HIP memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HIP memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]dstImgSizesdestination image sizes ( RpptImagePatchPtr type pointer to array, in pinned/HOST memory, of size batchSize)
    [in]interpolationTypeInterpolation type used in RpptInterpolationType
    [in]roiTensorPtrSrcROI data in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_resize_host()

RppStatus rppt_resize_host ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
RpptImagePatchPtr  dstImgSizes,
RpptInterpolationType  interpolationType,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Resize augmentation on HOST backend for a NCHW/NHWC layout tensor.

The resize augmentation performs an image resize for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HOST memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HOST memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]dstImgSizesdestination image sizes ( RpptImagePatchPtr type pointer to array, in HOST memory, of size batchSize)
    [in]interpolationTypeInterpolation type used in RpptInterpolationType
    [in]roiTensorPtrSrcROI data in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_resize_mirror_normalize_gpu()

RppStatus rppt_resize_mirror_normalize_gpu ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
RpptImagePatchPtr  dstImgSizes,
RpptInterpolationType  interpolationType,
Rpp32f meanTensor,
Rpp32f stdDevTensor,
Rpp32u mirrorTensor,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Resize mirror normalize augmentation on HIP backend for a NCHW/NHWC layout tensor.

The resize mirror normalize augmentation performs an image resize, an optional mirror and/or normalize, for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HIP memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HIP memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]dstImgSizesdestination image sizes ( RpptImagePatchPtr type pointer to array, in HIP memory, of size batchSize)
    [in]interpolationTypeInterpolation type used in RpptInterpolationType
    [in]meanTensormean value for each image in the batch (meanTensor[n] >= 0, 1D tensor in pinned/HOST memory, of size = batchSize for greyscale images, size = batchSize * 3 for RGB images))
    [in]stdDevTensorstandard deviation value for each image in the batch (stdDevTensor[n] >= 0, 1D tensor in pinned/HOST memory, of size = batchSize for greyscale images, size = batchSize * 3 for RGB images)
    [in]mirrorTensormirror flag value to set mirroring on/off (1D tensor in pinned/HOST memory, of size batchSize, with mirrorTensor[n] = 0/1)
    [in]roiTensorPtrSrcROI data in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_resize_mirror_normalize_host()

RppStatus rppt_resize_mirror_normalize_host ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
RpptImagePatchPtr  dstImgSizes,
RpptInterpolationType  interpolationType,
Rpp32f meanTensor,
Rpp32f stdDevTensor,
Rpp32u mirrorTensor,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Resize mirror normalize augmentation on HOST backend for a NCHW/NHWC layout tensor.

The resize mirror normalize augmentation performs an image resize, an optional mirror and/or normalize, for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HOST memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HOST memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]dstImgSizesdestination image sizes ( RpptImagePatchPtr type pointer to array, in HOST memory, of size batchSize)
    [in]interpolationTypeInterpolation type used in RpptInterpolationType
    [in]meanTensormean value for each image in the batch (meanTensor[n] >= 0, 1D tensor in HOST memory, of size = batchSize for greyscale images, size = batchSize * 3 for RGB images))
    [in]stdDevTensorstandard deviation value for each image in the batch (stdDevTensor[n] >= 0, 1D tensor in HOST memory, of size = batchSize for greyscale images, size = batchSize * 3 for RGB images)
    [in]mirrorTensormirror flag value to set mirroring on/off (1D tensor in HOST memory, of size batchSize, with mirrorTensor[n] = 0/1)
    [in]roiTensorPtrSrcROI data in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_rotate_gpu()

RppStatus rppt_rotate_gpu ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
Rpp32f angle,
RpptInterpolationType  interpolationType,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Rotate augmentation on HIP backend for a NCHW/NHWC layout tensor.

The rotate augmentation performs a rotate transformations for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HIP memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HIP memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]angleimage rotation angle in degrees - positive deg-anticlockwise/negative deg-clockwise (1D tensor in pinned/HOST memory, of size batchSize)
    [in]interpolationTypeInterpolation type used (RpptInterpolationType::XYWH or RpptRoiType::LTRB)
    [in]roiTensorPtrSrcROI data in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_rotate_host()

RppStatus rppt_rotate_host ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
Rpp32f angle,
RpptInterpolationType  interpolationType,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Rotate augmentation on HOST backend for a NCHW/NHWC layout tensor.

The rotate augmentation performs a rotate transformations for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HOST memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HOST memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]angleimage rotation angle in degrees - positive deg-anticlockwise/negative deg-clockwise (1D tensor in HOST memory, of size batchSize)
    [in]interpolationTypeInterpolation type used (RpptInterpolationType::XYWH or RpptRoiType::LTRB)
    [in]roiTensorPtrSrcROI data in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_slice_gpu()

RppStatus rppt_slice_gpu ( RppPtr_t  srcPtr,
RpptGenericDescPtr  srcGenericDescPtr,
RppPtr_t  dstPtr,
RpptGenericDescPtr  dstGenericDescPtr,
Rpp32s anchorTensor,
Rpp32s shapeTensor,
RppPtr_t  fillValue,
bool  enablePadding,
Rpp32u roiTensor,
rppHandle_t  rppHandle 
)

Slice augmentation GPU.

This function performs slice augmentation on a generic 4D tensor. Slice augmentation involves selecting a region of interest (ROI) from the source tensor and copying it to the destination tensor. Support added for f32 -> f32 and u8 -> u8 dataypes.

Parameters
[in]srcPtrsource tensor memory in HIP memory
[in]srcGenericDescPtrsource tensor descriptor
[out]dstPtrdestination tensor memory in HIP memory
[in]dstGenericDescPtrdestination tensor descriptor
[in]anchorTensorstarting index of the slice for each dimension in input (1D tensor in pinned/HOST memory of size = batchSize * numberOfDimensions)
[in]shapeTensorlength of the slice for each dimension in input (1D tensor in pinned/HOST memory of size = batchSize * numberOfDimensions)
[in]fillValuefill value that is used to fill output if enablePadding is set to true
[in]enablePaddingboolean flag to specify if padding is enabled or not
[in]roiTensorroi data in pinned/HOST memory (1D tensor of size = batchSize * numberOfDimensions * 2)
[in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
Returns
A RppStatus enumeration.
Return values
RPP_SUCCESSSuccessful completion.
RPP_ERROR*Unsuccessful completion.

◆ rppt_slice_host()

RppStatus rppt_slice_host ( RppPtr_t  srcPtr,
RpptGenericDescPtr  srcGenericDescPtr,
RppPtr_t  dstPtr,
RpptGenericDescPtr  dstGenericDescPtr,
Rpp32s anchorTensor,
Rpp32s shapeTensor,
RppPtr_t  fillValue,
bool  enablePadding,
Rpp32u roiTensor,
rppHandle_t  rppHandle 
)

Slice augmentation HOST.

This function performs slice augmentation on a generic 4D tensor. Slice augmentation involves selecting a region of interest (ROI) from the source tensor and copying it to the destination tensor. Support added for f32 -> f32 and u8 -> u8 dataypes.

Parameters
[in]srcPtrsource tensor memory in HOST memory
[in]srcGenericDescPtrsource tensor descriptor
[out]dstPtrdestination tensor memory in HOST memory
[in]dstGenericDescPtrdestination tensor descriptor
[in]anchorTensorstarting index of the slice for each dimension in input (1D tensor of size = batchSize * numberOfDimensions)
[in]shapeTensorlength of the slice for each dimension in input (1D tensor of size = batchSize * numberOfDimensions)
[in]fillValuefill value that is used to fill output if enablePadding is set to true
[in]enablePaddingboolean flag to specify if padding is enabled or not
[in]roiTensorroi data in HOST memory (1D tensor of size = batchSize * numberOfDimensions * 2)
[in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
Returns
A RppStatus enumeration.
Return values
RPP_SUCCESSSuccessful completion.
RPP_ERROR*Unsuccessful completion.

◆ rppt_transpose_gpu()

RppStatus rppt_transpose_gpu ( RppPtr_t  srcPtr,
RpptGenericDescPtr  srcGenericDescPtr,
RppPtr_t  dstPtr,
RpptGenericDescPtr  dstGenericDescPtr,
Rpp32u permTensor,
Rpp32u roiTensor,
rppHandle_t  rppHandle 
)

Transpose Generic augmentation on HIP backend.

The transpose augmentation performs an input-permutation based transpose on a generic ND Tensor.

Parameters
[in]srcPtrsource tensor in HIP memory
[in]srcGenericDescPtrsource tensor descriptor
[out]dstPtrsource tensor in HIP memory
[in]dstGenericDescPtrdestination tensor descriptor
[in]permTensorpermutation tensor for transpose operation in pinned memory
[in]roiTensorROI data for each element in source tensor (tensor of batchSize * number of dimensions * 2 values)
[in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
Returns
A RppStatus enumeration.
Return values
RPP_SUCCESSSuccessful completion.
RPP_ERROR*Unsuccessful completion.

◆ rppt_transpose_host()

RppStatus rppt_transpose_host ( RppPtr_t  srcPtr,
RpptGenericDescPtr  srcGenericDescPtr,
RppPtr_t  dstPtr,
RpptGenericDescPtr  dstGenericDescPtr,
Rpp32u permTensor,
Rpp32u roiTensor,
rppHandle_t  rppHandle 
)

Transpose Generic augmentation on HOST backend.

The transpose augmentation performs an input-permutation based transpose on a generic ND Tensor.

Parameters
[in]srcPtrsource tensor in HOST memory
[in]srcGenericDescPtrsource tensor descriptor
[out]dstPtrsource tensor in HOST memory
[in]dstGenericDescPtrdestination tensor descriptor
[in]permTensorpermutation tensor for transpose operation
[in]roiTensorROI data for each element in source tensor (tensor of batchSize * number of dimensions * 2 values)
[in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
Returns
A RppStatus enumeration.
Return values
RPP_SUCCESSSuccessful completion.
RPP_ERROR*Unsuccessful completion.

◆ rppt_warp_affine_gpu()

RppStatus rppt_warp_affine_gpu ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
Rpp32f affineTensor,
RpptInterpolationType  interpolationType,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Warp affine augmentation on HIP backend for a NCHW/NHWC layout tensor.

The warp affine augmentation performs affine transformations for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HIP memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HIP memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]affineTensoraffine matrix values for transformation calculation (2D tensor in pinned/HOST memory, of size batchSize * 6 for each image in batch)
    [in]interpolationTypeInterpolation type used (RpptInterpolationType::XYWH or RpptRoiType::LTRB)
    [in]roiTensorPtrSrcROI data in HIP memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HIP handle created with rppCreateWithStreamAndBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.

◆ rppt_warp_affine_host()

RppStatus rppt_warp_affine_host ( RppPtr_t  srcPtr,
RpptDescPtr  srcDescPtr,
RppPtr_t  dstPtr,
RpptDescPtr  dstDescPtr,
Rpp32f affineTensor,
RpptInterpolationType  interpolationType,
RpptROIPtr  roiTensorPtrSrc,
RpptRoiType  roiType,
rppHandle_t  rppHandle 
)

Warp affine augmentation on HOST backend for a NCHW/NHWC layout tensor.

The warp affine augmentation performs affine transformations for a batch of RGB(3 channel) / greyscale(1 channel) images with an NHWC/NCHW tensor layout.

  • srcPtr depth ranges - Rpp8u (0 to 255), Rpp16f (0 to 1), Rpp32f (0 to 1), Rpp8s (-128 to 127).
  • dstPtr depth ranges - Will be same depth as srcPtr.
    Sample Input
    Sample Output
    Parameters
    [in]srcPtrsource tensor in HOST memory
    [in]srcDescPtrsource tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3)
    [out]dstPtrdestination tensor in HOST memory
    [in]dstDescPtrdestination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr)
    [in]affineTensoraffine matrix values for transformation calculation (2D tensor in HOST memory, of size batchSize * 6 for each image in batch)
    [in]interpolationTypeInterpolation type used (RpptInterpolationType::XYWH or RpptRoiType::LTRB)
    [in]roiTensorPtrSrcROI data in HOST memory, for each image in source tensor (2D tensor of size batchSize * 4, in either format - XYWH(xy.x, xy.y, roiWidth, roiHeight) or LTRB(lt.x, lt.y, rb.x, rb.y))
    [in]roiTypeROI type used (RpptRoiType::XYWH or RpptRoiType::LTRB)
    [in]rppHandleRPP HOST handle created with rppCreateWithBatchSize()
    Returns
    A RppStatus enumeration.
    Return values
    RPP_SUCCESSSuccessful completion.
    RPP_ERROR*Unsuccessful completion.