[Feature] Added agnostic_nms onnx converter for tensorrt 8

[Jun0922]

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.

Motivation

Please describe the motivation for this PR and the goal you want to achieve through this PR.

I wanted my onnx model to convert with agnostic_nms algorithm, which tensorrt >= 8.6 supports and tested it works. You can test it with '--agnostic-nms' flag with 'export_onnx.py'. The feature is on official tensorrt and I just changed it to use the plugin during onnx convert. You can check the tensorrt feature below.

https://github.com/NVIDIA/TensorRT/blob/release/8.6/plugin/efficientNMSPlugin/EfficientNMSPlugin_PluginConfig.yaml

Modification

Please briefly describe what modification is made in this PR.

Before: It didn't support agnostic_nms with tensorrt8.
After: It supports agnostic_nms with tensorrt >= 8.6.

You can see the difference below. The first one is before agnostic_nms which means if model considers it has two classes, returns two bboxes for one object. The second one is after agnostic_nms which gives just one bbox that has the highest confidence score.

BEFORE

before-agnostic-nms.mp4

AFTER

after-agnostic-nms.mp4

Both of them served with NVIDIA Deepstream.

BC-breaking (Optional)

Does the modification introduce changes that break the backward compatibility of the downstream repos?
If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.

Use cases (Optional)

If this PR introduces a new feature, it is better to list some use cases here and update the documentation.

Checklist

1. Pre-commit or other linting tools are used to fix potential lint issues.
2. The modification is covered by complete unit tests. If not, please add more unit tests to ensure the correctness.
3. If the modification has a potential influence on downstream projects, this PR should be tested with downstream projects, like MMDetection or MMClassification.
4. The documentation has been modified accordingly, like docstring or example tutorials.

[CLAassistant]

All committers have signed the CLA.