Training&Testing dataset

[Tae Bok Lee]

Thanks for your kind work.

I have a few simple questions about training&testing dataset.
In your paper, dataet is mentioned as follows,

1. 300VW face dataset
   "Specifically, we select 83 videos as our training data and 9 videos as testing data from the 114 videos in the 300VW dataset."
2. VidTIMIT dataset
   "We randomly select 11 videos from VidTIMIT, synthesize the corresponding test-ing data, and directly run the different methods on them."

Can you provide the details of your training and testing set, please?
(the files or list of files used for training and testing..etc)
And, how do you get the bounding box of 300VW face dataset?
I look forward to hearing from you. Thank you.