Box filter padding updates by Srihari-mcw · Pull Request #597 · ROCm/rpp

Srihari-mcw · 2025-08-13T12:17:20Z

Adds nearest neighbors padding for box filter augmentation (Current ToT version had border pixels fading out, the PR attempts to fix the same )
Separated the compute functions of all bit depths on both HOST and HIP Backends (Avoids additional bit depth conversions)
Templated the compute functions for all filter sizes into one on the HIP Backend which improves/reduces lines of code significantly

… PLN3

src/include/common/cpu/rpp_cpu_filter.hpp

src/include/common/hip/rpp_hip_load_store.hpp

src/modules/tensor/rppt_tensor_filter_augmentations.cpp

rrawther

added comments

src/modules/tensor/cpu/kernel/box_filter.cpp

…adHorizontal

rrawther · 2025-09-03T23:23:35Z

@AryanSalmanpour : Can you please review the HIP code for this PR

rrawther

please resolve all conversations

LakshmiKumar23 · 2025-09-16T22:54:56Z

@spolifroni-amd can you fix the docs issue. This is blocking our merge

* Updates for box filter to test * Box Filter Newer Commit with fixes * Further updates to match PKD3 and PLN3 output * Fixes for RPP Box filter border replicate - Fix accuracy * Rename variables * Restore AVX and Update AVX code for box filter * Box Filter float updates * Updates to float version * Compilation fixes * HIP Updates for Box Filter kernelSize = 3 * HIP Updates for PLN Code * HIP Updates for kernelSize = 5 * HIP Updates for kernelSize = 7 * HIP Updates for kernelSize = 9 * Remove additional else * Fix issues with alignedLength for kernelSize 3 float variants PKD3 to PLN3 * Add golden outputs * Add additional borderType parameter * Separate float implementation for 3x3 box filter planar * Rename function * Introduce functions to calculate in float type itself * Updates for PKD variants * Compilation fixes * Updates for PKD3 to PLN3 * Fix accuracy issues * PLN3 to PKD3 updates for box filter * Float shared variables * Overload box filter for various kernelSizes * Template the PKD3 and PLN3 implementations * Template the rest of the implementations * Initial cleanup of HIP Code * Cleanup of code with parameter passing launch of kernel * Add further comments * Add more comments and minor cleanups further * Initial cleanup of cpu kernel code * Add comments for padding * Add reference outputs in docs/data/doxygenOutputs * Update documentation * Minor corrections * Updates to move filter structure as common code * Update the separation based on type * Update rpp_hip_load_store.hpp * Updates for separating i8 for PLN3 code * Fixes for PLN3 and further updates for kernelSize = 3 * I8 updates for kernelSize = 5 * Updates for kernelSize = 7 * Updates for kernelSize = 9 * Fixes for kernelSize = 9 * Remove saturate pixel * Rename function * Update comment * Add declarations for I8 functions with rounding * Whitespace and type updates * Updates for performance - U8/I8 * Templated the box filter compute functions for all kernel sizes and optimized the I8 variants * Update comments * Make initial changes to template unpack function * Fixes for box filter compilation * modified padding load logic * Update the round function used * Rename verticalDirection and horizontalDirection to padVertical and padHorizontal * Fix accuracy issues --------- Co-authored-by: Abishek <52214183+r-abishek@users.noreply.github.com> Co-authored-by: HazarathKumarM <hazarathkumar@multicorewareinc.com> Co-authored-by: Kiriti Gowda <kiritigowda@gmail.com> Co-authored-by: Lakshmi Kumar <lakshmi.kumar@amd.com>

* Box filter - padding updates (#597) * Updates for box filter to test * Box Filter Newer Commit with fixes * Further updates to match PKD3 and PLN3 output * Fixes for RPP Box filter border replicate - Fix accuracy * Rename variables * Restore AVX and Update AVX code for box filter * Box Filter float updates * Updates to float version * Compilation fixes * HIP Updates for Box Filter kernelSize = 3 * HIP Updates for PLN Code * HIP Updates for kernelSize = 5 * HIP Updates for kernelSize = 7 * HIP Updates for kernelSize = 9 * Remove additional else * Fix issues with alignedLength for kernelSize 3 float variants PKD3 to PLN3 * Add golden outputs * Add additional borderType parameter * Separate float implementation for 3x3 box filter planar * Rename function * Introduce functions to calculate in float type itself * Updates for PKD variants * Compilation fixes * Updates for PKD3 to PLN3 * Fix accuracy issues * PLN3 to PKD3 updates for box filter * Float shared variables * Overload box filter for various kernelSizes * Template the PKD3 and PLN3 implementations * Template the rest of the implementations * Initial cleanup of HIP Code * Cleanup of code with parameter passing launch of kernel * Add further comments * Add more comments and minor cleanups further * Initial cleanup of cpu kernel code * Add comments for padding * Add reference outputs in docs/data/doxygenOutputs * Update documentation * Minor corrections * Updates to move filter structure as common code * Update the separation based on type * Update rpp_hip_load_store.hpp * Updates for separating i8 for PLN3 code * Fixes for PLN3 and further updates for kernelSize = 3 * I8 updates for kernelSize = 5 * Updates for kernelSize = 7 * Updates for kernelSize = 9 * Fixes for kernelSize = 9 * Remove saturate pixel * Rename function * Update comment * Add declarations for I8 functions with rounding * Whitespace and type updates * Updates for performance - U8/I8 * Templated the box filter compute functions for all kernel sizes and optimized the I8 variants * Update comments * Make initial changes to template unpack function * Fixes for box filter compilation * modified padding load logic * Update the round function used * Rename verticalDirection and horizontalDirection to padVertical and padHorizontal * Fix accuracy issues --------- Co-authored-by: Abishek <52214183+r-abishek@users.noreply.github.com> Co-authored-by: HazarathKumarM <hazarathkumar@multicorewareinc.com> Co-authored-by: Kiriti Gowda <kiritigowda@gmail.com> Co-authored-by: Lakshmi Kumar <lakshmi.kumar@amd.com> * Docs - removing SP6 (#609) Co-authored-by: Kiriti Gowda <kiritigowda@gmail.com> * Packaging - Remove Meta Package dependency (#610) Co-authored-by: Kiriti Gowda <kiritigowda@gmail.com> * Bugfix - Handle (#600) * Fix handle mem leaks * Add variable to capture rpp::try_ return status * Add handle deletes inside lambda function * modified handle delete --------- Co-authored-by: Maddisetty <hmaddise@ctr2-alola-login-01.amd.com> Co-authored-by: Kiriti Gowda <kiritigowda@gmail.com> Co-authored-by: HazarathKumarM <hazarathkumar@multicorewareinc.com> * Docs - Updated doxyfile with correct api input path (#614) Co-authored-by: Kiriti Gowda <kiritigowda@gmail.com> * Test Package - Add cmake deps (#617) * Solarize augmentation - HOST and HIP (#605) * Solarize HIP and HOST implementation * cleanup the code and fix pkd3-pkd3 performance * Add golden output and doxygen comments * Add cheks for Threshold param * modified case num for solarize * minor fix * fix load/store calls * Address review comments * minor fix --------- Co-authored-by: HazarathKumarM <hazarathkumar@multicorewareinc.com> Co-authored-by: Maddisetty <hmaddise@ctr2-alola-login-01.amd.com> Co-authored-by: Kiriti Gowda <kiritigowda@gmail.com> * Docs - Bump rocm-docs-core[api_reference] from 1.23.0 to 1.24.1 in /docs/sphinx (#620) Bumps [rocm-docs-core[api_reference]](https://github.com/ROCm/rocm-docs-core) from 1.23.0 to 1.24.1. - [Release notes](https://github.com/ROCm/rocm-docs-core/releases) - [Changelog](https://github.com/ROCm/rocm-docs-core/blob/v1.24.1/CHANGELOG.md) - [Commits](ROCm/rocm-docs-core@v1.23.0...v1.24.1) --- updated-dependencies: - dependency-name: rocm-docs-core[api_reference] dependency-version: 1.24.1 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Docs - Bump rocm-docs-core[api_reference] from 1.24.1 to 1.25.0 in /docs/sphinx (#622) Bumps [rocm-docs-core[api_reference]](https://github.com/ROCm/rocm-docs-core) from 1.24.1 to 1.25.0. - [Release notes](https://github.com/ROCm/rocm-docs-core/releases) - [Changelog](https://github.com/ROCm/rocm-docs-core/blob/v1.25.0/CHANGELOG.md) - [Commits](ROCm/rocm-docs-core@v1.24.1...v1.25.0) --- updated-dependencies: - dependency-name: rocm-docs-core[api_reference] dependency-version: 1.25.0 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * Warning fixes for HIP Backend - Check return status (#606) * Initial hip memory fixes * Further hip warning fixes * Default build warning fixes - group III --------- Co-authored-by: Srihari-mcw <srihari@multicorewareinc.com> Co-authored-by: Kiriti Gowda <kiritigowda@gmail.com> * Docs - Bump rocm-docs-core[api_reference] from 1.25.0 to 1.26.0 in /docs/sphinx (#623) Bumps [rocm-docs-core[api_reference]](https://github.com/ROCm/rocm-docs-core) from 1.25.0 to 1.26.0. - [Release notes](https://github.com/ROCm/rocm-docs-core/releases) - [Changelog](https://github.com/ROCm/rocm-docs-core/blob/v1.26.0/CHANGELOG.md) - [Commits](ROCm/rocm-docs-core@v1.25.0...v1.26.0) --- updated-dependencies: - dependency-name: rocm-docs-core[api_reference] dependency-version: 1.26.0 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> * ROCm 7.1 - Changelog updates (#625) --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: Srihari-mcw <96763064+Srihari-mcw@users.noreply.github.com> Co-authored-by: Abishek <52214183+r-abishek@users.noreply.github.com> Co-authored-by: HazarathKumarM <hazarathkumar@multicorewareinc.com> Co-authored-by: Lakshmi Kumar <lakshmi.kumar@amd.com> Co-authored-by: spolifroni-amd <Sandra.Polifroni@amd.com> Co-authored-by: arvindcheru <90783369+arvindcheru@users.noreply.github.com> Co-authored-by: Maddisetty <hmaddise@ctr2-alola-login-01.amd.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: Srihari-mcw <srihari@multicorewareinc.com>

* Updates for box filter to test * Box Filter Newer Commit with fixes * Further updates to match PKD3 and PLN3 output * Fixes for RPP Box filter border replicate - Fix accuracy * Rename variables * Restore AVX and Update AVX code for box filter * Box Filter float updates * Updates to float version * Compilation fixes * HIP Updates for Box Filter kernelSize = 3 * HIP Updates for PLN Code * HIP Updates for kernelSize = 5 * HIP Updates for kernelSize = 7 * HIP Updates for kernelSize = 9 * Remove additional else * Fix issues with alignedLength for kernelSize 3 float variants PKD3 to PLN3 * Add golden outputs * Add additional borderType parameter * Separate float implementation for 3x3 box filter planar * Rename function * Introduce functions to calculate in float type itself * Updates for PKD variants * Compilation fixes * Updates for PKD3 to PLN3 * Fix accuracy issues * PLN3 to PKD3 updates for box filter * Float shared variables * Overload box filter for various kernelSizes * Template the PKD3 and PLN3 implementations * Template the rest of the implementations * Initial cleanup of HIP Code * Cleanup of code with parameter passing launch of kernel * Add further comments * Add more comments and minor cleanups further * Initial cleanup of cpu kernel code * Add comments for padding * Add reference outputs in docs/data/doxygenOutputs * Update documentation * Minor corrections * Updates to move filter structure as common code * Update the separation based on type * Update rpp_hip_load_store.hpp * Updates for separating i8 for PLN3 code * Fixes for PLN3 and further updates for kernelSize = 3 * I8 updates for kernelSize = 5 * Updates for kernelSize = 7 * Updates for kernelSize = 9 * Fixes for kernelSize = 9 * Remove saturate pixel * Rename function * Update comment * Add declarations for I8 functions with rounding * Whitespace and type updates * Updates for performance - U8/I8 * Templated the box filter compute functions for all kernel sizes and optimized the I8 variants * Update comments * Make initial changes to template unpack function * Fixes for box filter compilation * modified padding load logic * Update the round function used * Rename verticalDirection and horizontalDirection to padVertical and padHorizontal * Fix accuracy issues --------- Co-authored-by: Abishek <52214183+r-abishek@users.noreply.github.com> Co-authored-by: HazarathKumarM <hazarathkumar@multicorewareinc.com> Co-authored-by: Kiriti Gowda <kiritigowda@gmail.com> Co-authored-by: Lakshmi Kumar <lakshmi.kumar@amd.com>

Srihari-mcw added 30 commits June 16, 2025 16:20

Updates for box filter to test

e2f34fb

Box Filter Newer Commit with fixes

a1218c3

Further updates to match PKD3 and PLN3 output

5bec7e1

Fixes for RPP Box filter border replicate - Fix accuracy

c7e925b

Rename variables

ff0582d

Restore AVX and Update AVX code for box filter

a6b4afd

Box Filter float updates

4c34bd7

Updates to float version

92f338c

Compilation fixes

db2f2ba

HIP Updates for Box Filter kernelSize = 3

09ee499

HIP Updates for PLN Code

1c774d6

HIP Updates for kernelSize = 5

35aa810

HIP Updates for kernelSize = 7

edd3dc3

HIP Updates for kernelSize = 9

f6e48c9

Remove additional else

dc7ef37

Fix issues with alignedLength for kernelSize 3 float variants PKD3 to…

27dfedc

… PLN3

Add golden outputs

17d597e

Add additional borderType parameter

46ae0ea

Separate float implementation for 3x3 box filter planar

32cb4be

Rename function

6abfcc7

Introduce functions to calculate in float type itself

0fbc9fe

Updates for PKD variants

03fdbe6

Compilation fixes

22d8f5a

Updates for PKD3 to PLN3

52a7cef

Fix accuracy issues

4d75b0d

PLN3 to PKD3 updates for box filter

7f456d7

Float shared variables

747f4b3

Overload box filter for various kernelSizes

6fb913f

Template the PKD3 and PLN3 implementations

6e77dbb

Template the rest of the implementations

c2c42ff

kiritigowda and others added 4 commits August 13, 2025 11:28

Merge branch 'develop' into box_filter_padding_updates

eebd365

Merge branch 'develop' into box_filter_padding_updates

3710940

Merge branch 'develop' into box_filter_padding_updates

2f14374

Merge branch 'develop' into box_filter_padding_updates

6babc28

rrawther reviewed Aug 23, 2025

View reviewed changes

src/include/common/cpu/rpp_cpu_filter.hpp Show resolved Hide resolved

rrawther reviewed Aug 23, 2025

View reviewed changes

src/include/common/hip/rpp_hip_load_store.hpp Show resolved Hide resolved

rrawther reviewed Aug 23, 2025

View reviewed changes

src/include/common/hip/rpp_hip_load_store.hpp Outdated Show resolved Hide resolved

Srihari-mcw and others added 3 commits August 26, 2025 16:59

Merge branch 'develop' into box_filter_padding_updates

f19c3fd

Update the round function used

2d47a76

Merge branch 'develop' into box_filter_padding_updates

c89f6f5

rrawther reviewed Aug 27, 2025

View reviewed changes

src/include/common/hip/rpp_hip_load_store.hpp Show resolved Hide resolved

rrawther reviewed Aug 27, 2025

View reviewed changes

src/modules/tensor/rppt_tensor_filter_augmentations.cpp Show resolved Hide resolved

rrawther requested changes Aug 28, 2025

View reviewed changes

src/modules/tensor/cpu/kernel/box_filter.cpp Outdated Show resolved Hide resolved

Srihari-mcw added 3 commits September 1, 2025 10:40

Rename verticalDirection and horizontalDirection to padVertical and p…

dc863ef

…adHorizontal

Merge branch 'develop' into box_filter_padding_updates

0de01f8

Fix accuracy issues

51c3cfe

rrawther requested a review from AryanSalmanpour September 3, 2025 23:23

rrawther requested changes Sep 4, 2025

View reviewed changes

rrawther approved these changes Sep 11, 2025

View reviewed changes

Merge branch 'develop' into box_filter_padding_updates

d4d1bb8

AryanSalmanpour approved these changes Sep 16, 2025

View reviewed changes

kiritigowda merged commit 89de11e into ROCm:develop Sep 16, 2025
4 of 5 checks passed

Srihari-mcw mentioned this pull request Sep 17, 2025

Removal of Integer Pack Load Functions r-abishek/rpp#487

Merged

SundarRajan28 mentioned this pull request Sep 18, 2025

Update blur vx_rpp extension ROCm/MIVisionX#1557

Merged

1 task

r-abishek mentioned this pull request Sep 24, 2025

Remove unnecessary int hip load/stores #621

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Box filter padding updates#597

Box filter padding updates#597
kiritigowda merged 70 commits intoROCm:developfrom
Srihari-mcw:box_filter_padding_updates

Srihari-mcw commented Aug 13, 2025 •

edited

Loading

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

rrawther left a comment

Uh oh!

Uh oh!

rrawther commented Sep 3, 2025

Uh oh!

rrawther left a comment

Uh oh!

LakshmiKumar23 commented Sep 16, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

Conversation

Srihari-mcw commented Aug 13, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

rrawther left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

rrawther commented Sep 3, 2025

Uh oh!

rrawther left a comment

Choose a reason for hiding this comment

Uh oh!

LakshmiKumar23 commented Sep 16, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

Srihari-mcw commented Aug 13, 2025 •

edited

Loading