PaddleOCR c++ infer

Build the project using VS2019 and cmake on win10.

Environment

CUDA:11.3

TensorRT:8.4.0.6

OpenCV:4.8.0

Run the main.cpp

After downloading the model, modify the model path in main.cpp.

std::string model_dir = "D:/code/PaddleOCR/model/";
FLAGS_det_model_dir = model_dir + "ch_PP-OCRv4_det_server_infer";
FLAGS_rec_model_dir = model_dir + "ch_PP-OCRv4_rec_server_infer";
// image path
FLAGS_image_dir = "D:/12.jpg";

Acknowledgement

1.PaddleOCR