We have provided examples of testing our method in the README.md, which involves inference on the test dataset. You can modify it according to your needs or test it on other datasets.
The inference speed of our method depends on the underlying model we use. Our method uses lightweight convolutional models in both stages, thus possessing the ability for real-time inference. If you have higher real-time inference requirements, you can also change midnet/model.py in the code, just ensure that the input and output shapes of the model are consistent with the original.
If you have any other questions, please feel free to contact us.
Hello, thank you for your interest in our work.
We have provided examples of testing our method in the
README.md, which involves inference on the test dataset. You can modify it according to your needs or test it on other datasets.The inference speed of our method depends on the underlying model we use. Our method uses lightweight convolutional models in both stages, thus possessing the ability for real-time inference. If you have higher real-time inference requirements, you can also change
midnet/model.pyin the code, just ensure that the input and output shapes of the model are consistent with the original.If you have any other questions, please feel free to contact us.