DefTruth

add TNN C++ Demo link

[yolop.cpp](https://github.com/DefTruth/lite.ai/blob/main/ort/cv/yolop.cpp)

can not tar -xvf the pretrained model.

I have reimplement FaceBoxes with c++, [faceboxes.cpp](https://github.com/DefTruth/lite.ai/blob/main/lite/ort/cv/faceboxes.cpp). It seems FaceBoxes suitable for mobile device with some simple frontal faces, but can not get good accuracy for complicated situation.

add ONNXRuntime/MNN/TNN/NCNN C++ demo

2

add ONNXRuntime/MNN/TNN/NCNN C++ demo, the C++ source codes are listed as below: * [ONNXRuntime C++ MODNet](https://github.com/DefTruth/lite.ai.toolkit/blob/main/lite/ort/cv/modnet.cpp) * [MNN C++ MODNet](https://github.com/DefTruth/lite.ai.toolkit/blob/main/lite/mnn/cv/mnn_modnet.cpp) * [NCNN C++ MODNet](https://github.com/DefTruth/lite.ai.toolkit/blob/main/lite/ncnn/cv/ncnn_modnet.cpp) * [TNN C++ MODNet](https://github.com/DefTruth/lite.ai.toolkit/blob/main/lite/tnn/cv/tnn_modnet.cpp)

鉴于问这个问题的人比较多，我写了一份详细的知乎教程，请参考： * [🔧填坑: RobustVideoMatting(5k+🔥 star)视频抠图静态ONNX模型转换](https://zhuanlan.zhihu.com/p/459088407)

### PR types(PR类型) Other (Android) ### Describe - [x] bind SegmentationResult - [x] add `NewCxxRuntimeOption` method to convert Java RuntimeOption to Cxx RuntimeOption - [x] add `AllocateCxxResultFromJava`、`AllocateJavaResultFromCxx` and `NewJavaResultFromCxx` -...

### PR types(PR类型) Other ### Describe - Update ppseg jni via new api and optimize jni vis func

### PR types(PR类型) FastTokenizer ### Describe - [x] 支持FastDeploy+FastTokenizer Android交叉编译 - [x] UIE Android Example 测试，fp32 & fp16

🐛 [Bug] torch._export.verifier.SpecViolationError when using Torch-TensorRT

4

## Bug Description got the export error with torch_tensorrt 2.2.0 use dynamo IR ```bash torch._export.verifier.SpecViolationError: Node.meta max_pool2d_default is missing val field. ``` ## To Reproduce ```python optimized_model = torch_tensorrt.compile( self.model,...