Could you please kindly provide me with a training demo? I have downloaded the CREMA (audio/video) dataset you processed, but I have no idea what your directory structure is.
In addition, can I complete a simple training without using landmarks, using only the CREMA (audio/video) dataset and the /diffused-heads/ckpts/audio_encoder.pt you provided?
Could you please kindly provide me with a training demo? I have downloaded the CREMA (audio/video) dataset you processed, but I have no idea what your directory structure is.
In addition, can I complete a simple training without using landmarks, using only the CREMA (audio/video) dataset and the /diffused-heads/ckpts/audio_encoder.pt you provided?