search

cfeng16 / audio-visual-forensics

https://cfeng16.github.io/audio-visual-forensics/
MIT License
64 stars 5 forks source link

Pre-trained audio and video encoders #4

Closed adrianSRoman closed 5 months ago

adrianSRoman commented 7 months ago

Hi Chao,

First of all, thank you so much for sharing your research and making the code publicly available.

I would like to run some experiments with the model. However, I am running into problems finding the pre-trained Resnet18 2D+3D encoder, and the pre-trained VGGM audio encoder. I was wondering if you could share the pre-trained encoders? It seems like the Resnet18 you used has some slight variations compared to https://github.com/TengdaHan/DPC/blob/master/backbone/resnet_2d3d.py, and your VGGM compared to https://github.com/Cadene/pretrained-models.pytorch/blob/master/pretrainedmodels/models/vggm.py, so I want to make sure I use the right pre-trained model. Thanks in advance!

cfeng16 commented 5 months ago

I have updated the codebase. Please let me know if you have any questions