GeWu-Lab / MMCosine_ICASSP23

The code repo for ICASSP 2023 Paper "MMCosine: Multi-Modal Cosine Loss Towards Balanced Audio-Visual Fine-Grained Learning"
17 stars 1 forks source link

visual in CREMA-D dataset #2

Open PeiwenSun2000 opened 1 year ago

PeiwenSun2000 commented 1 year ago

Have you done any cropping or other pre-processing of the video frames in this dataset? My replication shows no change in the accuracy of the images during training.

Rick-Xu315 commented 1 year ago

Thanks for your interest in our work. We conduct the same preprocessing from OGM-GE python data/CREMAD/video_preprecessing.py.