search

SydCaption / SAAT

MIT License
62 stars 21 forks source link

About regional Dataset #20

Closed Dorothylyly closed 4 years ago

Dorothylyly commented 4 years ago
  1. msrvtt_roi_feat.h5
  2. msrvtt_roi_box.h5

this two h5 files have 10000 datasets respectively。 you just have 6513 videos ，but why you have 10000 datasets each file ？ what is the meaning of each dataset ? is it for frames ？

SydCaption commented 4 years ago

Hi, the msvtt dataset actually contains 10k videos, as can be found on http://ms-multimedia-challenge.com/2016/dataset. Each item in the .h5 file is the features or coordinate information of object proposals extracted from the centre frame of the input video.

Dorothylyly commented 4 years ago

thank you for your reply!! do you mean that : although msvtt contains 10k videos, you just use 6513 videos of it in your code ??

SydCaption commented 4 years ago

I'm sorry, but what do you mean by using 6513 videos? The official splits are 6513 in training set, 497 for validation and 2990 for test.

Dorothylyly commented 4 years ago

ooooooooooooooooh. It is my mistake, now I know it. thank you !!!!!!!!!!!!!!!!!!!!

Dorothylyly commented 4 years ago

I am a novice, I forgot that there are test set and validation set