Skip to content

G-API: Use different devices processing in streaming pipeline #21716

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
wants to merge 6 commits into from
Closed

G-API: Use different devices processing in streaming pipeline #21716

wants to merge 6 commits into from

Conversation

sivanov-work
Copy link
Contributor

@sivanov-work sivanov-work commented Mar 14, 2022 •
edited
Loading

Pull Request Readiness Checklist

See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request

  • I agree to contribute to the project under Apache 2 License.
  • To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
  • The PR is proposed to the proper branch
  • There is a reference to the original bug report and related work
  • There is accuracy test, performance test and test data in opencv_extra repository, if applicable
    Patch to opencv_extra has the same branch name.
  • The feature is well documented and sample code can be built with the project CMake

Build Configuration

force_builders=XCustom,Custom Win,Custom Mac
build_gapi_standalone:Linux x64=ade-0.1.1f
build_gapi_standalone:Win64=ade-0.1.1f
build_gapi_standalone:Mac=ade-0.1.1f
build_gapi_standalone:Linux x64 Debug=ade-0.1.1f

Xbuild_image:Custom=centos:7
Xbuildworker:Custom=linux-1
build_gapi_standalone:Custom=ade-0.1.1f

build_image:Custom=ubuntu-openvino-2021.3.0:20.04
Xbuild_image:Custom Win=openvino-2021.2.0
build_image:Custom Mac=openvino-2021.2.0

test_modules:Custom=gapi,python2,python3,java
test_modules:Custom Win=gapi,python2,python3,java
test_modules:Custom Mac=gapi,python2,python3,java

buildworker:Custom=linux-1
test_opencl:Custom=OFF
test_bigdata:Custom=1
test_filter:Custom=*

build_image:Custom Win=gapi-onevpl-2021.6.0
buildworker:Custom Win=windows-3
build_contrib:Custom Win=OFF

@sivanov-work sivanov-work force-pushed the vpl_different_devices branch from 9ecc0a6 to 54552f2 Compare April 1, 2022 11:02
@sivanov-work sivanov-work requested review from rgarnov, dmatveev and TolyaTalamanov and removed request for rgarnov and dmatveev April 4, 2022 13:13
@sivanov-work sivanov-work marked this pull request as ready for review April 4, 2022 13:13
rgarnov
rgarnov requested changes Apr 4, 2022
View reviewed changes
modules/gapi/samples/onevpl_infer_single_roi.cpp Outdated
accel_ctx.value());
std::cout << "enforce VPP preprocessing on " << device_id << std::endl;
// Turn on VPP PreprocesingEngine if available & requested
if (flow_settings->ie_preproc_enable) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ie_preproc_enable name can be misleading here, default preprocessing in IE is also ie_preproc

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ok,
ie_preproc_enable -> vpl_preproc_enable is suitable?

modules/gapi/samples/onevpl_infer_single_roi.cpp Outdated

// NB: consider NV12 surface because it's one of native GPU image format
face_net.pluginConfig({{"GPU_NV12_TWO_INPUTS", "YES" }});
std::cout << "enfore InferenceEngine NV12 blob" << std::endl;
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

enforce

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ok

modules/gapi/samples/onevpl_infer_single_roi.cpp Outdated
face_net.cfgContextParams(ctx_config);
std::cout << "enfore InferenceEngine remote context on device: " << device_id << std::endl;
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

enforce

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

thanks

modules/gapi/src/backends/ie/giebackend.cpp Outdated
@@ -712,6 +713,10 @@ inline IE::Blob::Ptr extractRemoteBlob(IECallContext& ctx, std::size_t i,
cv::MediaFrame frame = ctx.inFrame(i);
if (ctx.uu.preproc_engine_impl) {
GAPI_LOG_DEBUG(nullptr, "Try to use preprocessing for decoded remote frame in remote ctx");

//TODO
frame.blobParams();
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Todo?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Oh, looks like there is some exception will be thrown if there is a CPU adapter inside

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think I will change this block a little bit

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

extractRemoteBlob has gone away at all. please take a look

modules/gapi/samples/onevpl_infer_single_roi.cpp
// - pass such wrappers as constructor arguments for each component in pipeline:
// a) use special constructor for `onevpl::GSource`
// b) use `cfgContextParams` method of `cv::gapi::ie::Params` to charge PreprocesingEngine
// c) use `InferenceEngine::ParamMap` to activate remote ctx in Inference Engine
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks like it will be very nice to have an opportunity to configure everything in one place (set device/context only once) if user want everything to run on GPU for example. But can't find a way how to make it, at first glance

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Indeed, that's because we discussed DeviceSelector approach many times before

@sivanov-work sivanov-work requested a review from rgarnov April 5, 2022 07:50
@sivanov-work
Copy link
Contributor Author

Feature implemented here #22212

Closed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@TolyaTalamanov TolyaTalamanov Awaiting requested review from TolyaTalamanov

@dmatveev dmatveev Awaiting requested review from dmatveev

@rgarnov rgarnov Awaiting requested review from rgarnov

@mpashchenkov mpashchenkov Awaiting requested review from mpashchenkov

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@sivanov-work @rgarnov