rjhowell44

@youngjae-avikus the only thing you would add to your Pipeline is the Splitter. Then create two branches, each with a pgie and tracker and overlay sink. Then add the branches...

@youngjae-avikus using multiple Pipelines, each with their own Interpipe Source, PGIE, Tracker, OSD, and Window Sink, both listening to the same Interpipe Sink is a better way to go. Also,...

@youngjae-avikus I've pushed my preliminary work to a new branch [nms](https://github.com/prominenceai/deepstream-services-library/tree/nms_pph). The updated cpp example is at [non_maximum_suppression.cpp](https://github.com/prominenceai/deepstream-services-library/blob/nms_pph/examples/cpp/non_maximum_suppression.cpp). I've gotten as far as building you code into the new NMS-PPH...

@youngjae-avikus the above API definition has bee refactored to support both NMS and NMM

@youngjae-avikus I've pushed new updates to the [nms_pph](https://github.com/prominenceai/deepstream-services-library/tree/nms_pph) branch. Still a work in progress, but the basic functionality is there. Still needs more refactoring so I can add more unit...

@youngjae-avikus new updates have been pushed. Just for your information [API Tests](https://github.com/prominenceai/deepstream-services-library/blob/nms_pph/test/nmp/DslPphNmpApiTest.cpp) are done. [Unit Tests](https://github.com/prominenceai/deepstream-services-library/blob/nms_pph/test/nmp/DslPphNmpUnitTest.cpp) are still a work in progress.

@youngjae-avikus you can run the test cases as follows. API ``` ./dsl-test-app.exe -s [pph-nmp-api] ``` Unit ``` ./dsl-test-app.exe -s [NmsPph] ```

@youngjae-avikus currently, the `class agnostic option` is controlled by the `label_file` parameter. Just set the parameter to NULL for class agnostic operation. From the API comments above ```C * @param[in]...

@youngjae-avikus yes, for non agnostic operation, you use the same `label_file` as passed to the infer plugin via the config file

@youngjae-avikus ... although not complete, I've merged this work into the v0.26.alpha branch so that others can review/test. I have the API documented.... the [introduction](https://github.com/prominenceai/deepstream-services-library/blob/v0.26.alpha/docs/api-pph.md#non-maximum-processor-nmp-pad-probe-handler) still needs work though.