Gray Dove

The project cannot work with RTX3090 and CUDA11.1, considering the compatibility of 30 series and cuda11. Now, I am trying to buid the source in my work station.

Could you tell me which version of implementation for calculating the expression error?

1

There are two implementations (an official tensorflow version and a pytorch version) of the following paper: Y. Deng, J. Yang, S. Xu, D. Chen, Y. Jia, and X. Tong, Accurate...

Would you give me an advice of 'masks' of input images?

1

I have trained a model, named 'PCL-I2G-FF256-32frames-Modified-5e-5'. When I run the test.py with this setting below: ``` /public/user/anaconda3/envs/PCL-I2G/bin/python3.7 test.py --which_epoch bestval --gpu_ids 0 --partition test --average_mode after_softmax --topn 100 --force_redo...