RPPT Tensor Operations - Data Exchange Operations.#
RPPT Tensor Operations - Data Exchange Operations. More...
Functions | |
| RppStatus | rppt_copy (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, rppHandle_t rppHandle, RppBackend executionBackend) |
| Copy operation on HIP/HOST backend for a NCHW/NHWC layout tensor. More... | |
| RppStatus | rppt_channel_permute (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32u *permutationTensor, rppHandle_t rppHandle, RppBackend executionBackend) |
| Channel permute operation on HIP/HOST backend for a NCHW/NHWC layout tensor. More... | |
| RppStatus | rppt_color_to_greyscale (RppPtr_t srcPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, RpptSubpixelLayout srcSubpixelLayout, rppHandle_t rppHandle, RppBackend executionBackend) |
| Color to greyscale operation on HIP/HOST backend for a NCHW/NHWC layout tensor. More... | |
| RppStatus | rppt_yuv_to_rgb (RppPtr_t srcYPtr, RppPtr_t srcUVPtr, RpptDescPtr srcDescPtr, RppPtr_t dstPtr, RpptDescPtr dstDescPtr, Rpp32u src_y_pitch, Rpp32u src_uv_pitch, Rpp32u dst_pitch, Rpp32u width, Rpp32u height, RpptColorStandard col_standard, RpptColorRange color_range, rppHandle_t rppHandle, RppBackend executionBackend) |
| YUV to RGB color conversion on HIP backend (NV12 8-bit only) More... | |
Detailed Description
RPPT Tensor Operations - Data Exchange Operations.
Function Documentation
◆ rppt_channel_permute()
| RppStatus rppt_channel_permute | ( | RppPtr_t | srcPtr, |
| RpptDescPtr | srcDescPtr, | ||
| RppPtr_t | dstPtr, | ||
| RpptDescPtr | dstDescPtr, | ||
| Rpp32u * | permutationTensor, | ||
| rppHandle_t | rppHandle, | ||
| RppBackend | executionBackend | ||
| ) |
Channel permute operation on HIP/HOST backend for a NCHW/NHWC layout tensor.
This function performs one of six possible channel permutations (R-G-B, R-B-G, G-R-B, G-B-R, B-R-G, B-G-R) for an image in a batch of RGB(3 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] srcPtr source tensor in HIP memory (for HIP backend) or HOST memory (for HOST backend) [in] srcDescPtr source tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 3) [out] dstPtr destination tensor in HIP memory (for HIP backend) or HOST memory (for HOST backend) [in] dstDescPtr destination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr) [in] permutationTensor A tensor in HIP memory (for HIP backend) or HOST memory (for HOST backend) specifying the channel permutation for each image. Size: 3 × srcDescPtr->n. Each value must satisfy: 0 ≤ permutationTensor[i] ≤ 2. [in] rppHandle RPP HIP/HOST handle created with rppCreate()[in] executionBackend backend for execution (RppBackend::RPP_HOST_BACKEND or RppBackend::RPP_HIP_BACKEND)
- Returns
- A
RppStatusenumeration.
- Return values
-
RPP_SUCCESS Successful completion. RPP_ERROR* Unsuccessful completion.
◆ rppt_color_to_greyscale()
| RppStatus rppt_color_to_greyscale | ( | RppPtr_t | srcPtr, |
| RpptDescPtr | srcDescPtr, | ||
| RppPtr_t | dstPtr, | ||
| RpptDescPtr | dstDescPtr, | ||
| RpptSubpixelLayout | srcSubpixelLayout, | ||
| rppHandle_t | rppHandle, | ||
| RppBackend | executionBackend | ||
| ) |
Color to greyscale operation on HIP/HOST backend for a NCHW/NHWC layout tensor.
The color to greyscale operation runs for a batch of RGB(3 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] srcPtr source tensor in HIP memory (for HIP backend) or HOST memory (for HOST backend) [in] srcDescPtr source tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 3) [out] dstPtr destination tensor in HIP memory (for HIP backend) or HOST memory (for HOST backend) [in] dstDescPtr destination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr) [in] srcSubpixelLayout A RpptSubpixelLayout type enum to specify source subpixel layout (RGBtype or BGRtype) [in] rppHandle RPP HIP/HOST handle created with rppCreate()[in] executionBackend backend for execution (RppBackend::RPP_HOST_BACKEND or RppBackend::RPP_HIP_BACKEND)
- Returns
- A
RppStatusenumeration.
- Return values
-
RPP_SUCCESS Successful completion. RPP_ERROR* Unsuccessful completion.
◆ rppt_copy()
| RppStatus rppt_copy | ( | RppPtr_t | srcPtr, |
| RpptDescPtr | srcDescPtr, | ||
| RppPtr_t | dstPtr, | ||
| RpptDescPtr | dstDescPtr, | ||
| rppHandle_t | rppHandle, | ||
| RppBackend | executionBackend | ||
| ) |
Copy operation on HIP/HOST backend for a NCHW/NHWC layout tensor.
The copy operation runs a buffer copy 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] srcPtr source tensor in HIP memory (for HIP backend) or HOST memory (for HOST backend) [in] srcDescPtr source tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = 1/3) [out] dstPtr destination tensor in HIP memory (for HIP backend) or HOST memory (for HOST backend) [in] dstDescPtr destination tensor descriptor (Restrictions - numDims = 4, offsetInBytes >= 0, dataType = U8/F16/F32/I8, layout = NCHW/NHWC, c = same as that of srcDescPtr) [in] rppHandle RPP HIP/HOST handle created with rppCreate()[in] executionBackend backend for execution (RppBackend::RPP_HOST_BACKEND or RppBackend::RPP_HIP_BACKEND)
- Returns
- A
RppStatusenumeration.
- Return values
-
RPP_SUCCESS Successful completion. RPP_ERROR* Unsuccessful completion.
◆ rppt_yuv_to_rgb()
| RppStatus rppt_yuv_to_rgb | ( | RppPtr_t | srcYPtr, |
| RppPtr_t | srcUVPtr, | ||
| RpptDescPtr | srcDescPtr, | ||
| RppPtr_t | dstPtr, | ||
| RpptDescPtr | dstDescPtr, | ||
| Rpp32u | src_y_pitch, | ||
| Rpp32u | src_uv_pitch, | ||
| Rpp32u | dst_pitch, | ||
| Rpp32u | width, | ||
| Rpp32u | height, | ||
| RpptColorStandard | col_standard, | ||
| RpptColorRange | color_range, | ||
| rppHandle_t | rppHandle, | ||
| RppBackend | executionBackend | ||
| ) |
YUV to RGB color conversion on HIP backend (NV12 8-bit only)
Converts semi-planar NV12 (separate Y and interleaved UV planes) to packed RGB24.
- Source: srcYPtr = luma plane, srcUVPtr = interleaved UV
- src_y_pitch / src_uv_pitch: row strides in bytes
- Destination: packed RGB at dstPtr (RGB24, Rpp8u) with row stride dst_pitch bytes.
- Supported execution backend: RppBackend::RPP_HIP_BACKEND only.
- Parameters
-
[in] srcYPtr pointer to Y plane in HIP memory [in] srcUVPtr pointer to interleaved UV plane in HIP memory [in] srcDescPtr source tensor descriptor (dataType must be U8) [out] dstPtr pointer to RGB output buffer in HIP memory [in] dstDescPtr destination tensor descriptor (dataType must be U8) [in] src_y_pitch row pitch of Y plane in bytes [in] src_uv_pitch row pitch of UV plane in bytes [in] dst_pitch row pitch of RGB output in bytes [in] width image width in pixels [in] height image height in pixels [in] col_standard Luma/matrix family: RpptColorStandard (unknown values use BT.709). [in] color_range Luma range: RpptColorRange_STUDIO or RpptColorRange_FULL (other values behave like studio). [in] rppHandle RPP HIP handle created with rppCreate()[in] executionBackend must be RppBackend::RPP_HIP_BACKEND
- Returns
- A
RppStatusenumeration.
- Return values
-
RPP_SUCCESS Successful completion. RPP_ERROR* Unsuccessful completion (e.g. RPP_ERROR_NOT_IMPLEMENTED if executionBackend is not HIP).