YingqingHe / LVDM

LVDM: Latent Video Diffusion Models for High-Fidelity Long Video Generation
https://yingqinghe.github.io/LVDM/
MIT License
459 stars 18 forks source link

Sky_lapse dataset structure #7

Open Instassa opened 1 year ago

Instassa commented 1 year ago

Sky_lapse dataset download is being interrupted by the Google Drive for me, I guess it is size/speed limitations. Could someone please share the information about the structure of the dataset and what the format ucfTrainTestlist/classInd.txt?

xiaomingxige commented 1 year ago

~3{M@P8492C8)JCNWV4W8 T

kkedich commented 8 months ago

The Sky lapse dataset structure follows the second-level directory defined here. So, is:

dataset
    train
        videoxxx
            videoxxx_1
                 frame1.jpg
                 frame2.jpg
             videoxxx_2
                  frame1.jpg
             ...
        videoxxy
    test
        ...

And the ucfTrainTestlist/classInd.txt which can be download from here is a file with the label and class names of the UCF101 dataset:

1 ApplyEyeMakeup
2 ApplyLipstick
3 Archery
4 BabyCrawling
5 BalanceBeam
6 BandMarching