denfed / heartheflow Goto Github PK

Hi, thanks for your contribution to audio source localization.

Now, I am trying to train your model with VGGSound.
I have the VGGSound dataset but I do not know how to do preprocessing the dataset

It would be very appreciated if you share the preprocessing codes for VGGSound.

Thank you.

Hey, bro @denfed Recently I start to do researches with Sound Source Localization and I have get the SoundNet-Flickr test set. However I can't obtain the VGG-SS, the steps in vgg's paper is tedious. So could you share that with me? Best wishes to you if you will, thanks!

Hi! Encountered issues where no output data is saved in the outputdirectory/frames. Is there anything wrong with the training dataset that I downloaded? Thanks!

Hi! Your work is great. And could you please tell which tool do you use for the visualization of the extracted flow show. in Figure. 3 in you paper?

I use the flow_vis package for visualization but the visualization is not so clear to show the shape of the target object