search

Hangz-nju-cuhk / Talking-Face-Generation-DAVS

Code for Talking Face Generation by Adversarially Disentangled Audio-Visual Representation (AAAI 2019)
MIT License
819 stars 173 forks source link

Pre-Processing Data #47

Closed MrAsimZahid closed 4 years ago

MrAsimZahid commented 4 years ago

Hey @Hangz-nju-cuhk @liuziwei7 @liuyuisanai! I am trying to understand and reproduce the results of this repository end-to-end, So I could create a docker file and contribute. I have already read the paper thoroughly and analyzing the code now. But I am having a problem with pre-processing the data. Could you please guide how could I do this? Step by step process. Looking forward to you. Thank you.

Hangz-nju-cuhk commented 4 years ago

@MrAsimZahid Hi, you have to specify your problem before we can help you.

MrAsimZahid commented 4 years ago

@Hangz-nju-cuhk Let's say I have the "VoxCeleb" dataset How could I prepare it for training? By understanding your example I would be able to understand This folder code. How to use it?

Plus do you have access of LRW dataset? I am unable to gain access of it.

Hangz-nju-cuhk commented 4 years ago

@MrAsimZahid Have you read our paper carefully? Cause we state clearly that we do not use Voxceleb for training. The access to LRW dataset can be granted by VGG group and BBC only. Please refer to the dataset homepage.

MrAsimZahid commented 4 years ago

@Hangz-nju-cuhk I have already read the paper. And I wrote to BBC for LRW dataset access few days ago. They have declined it. As they stated: bbc screenshot

MrAsimZahid commented 4 years ago

Issue resolved. I will soon write an article about the whole project. Thanks @Hangz-nju-cuhk for helping out.