Skip to content

luqsthunder/Skelibras

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 

Repository files navigation

Skelibras - provisional website

Skelibras: a large annotated skeleton dataset of dynamic Libras signs comprised of unsegmented and segmented conversations.

Dataset organiztion

The directory structure is simple with each folder following the name of the sign. And each sign follows this nomenclature below:

  • VideoID---SignName---BeginInMiliseconds---EndInMiliSeconds---View---SignalyID.csv

The signs names follow the guia corpus guide:

Guia Corpus

All CSVs have the keys corresponding to the extracted joints, hand and body. All indices correspond to frames. And each joint is a string from a numpy.array, which can be mapped back to numpy.array as follows:

def parse_npy_vec_str(str_array_like):
    if not isinstance(str_array_like, str):
        return str_array_like

    res = str_array_like[1:len(str_array_like) - 1].split(' ')
    recovered_np_array = []
    for r in res:
        if r == '' or ' ' in r:
            continue
        f = float(r)
        recovered_np_array.append(f)
    return np.array(recovered_np_array)
        
sign = sign.applymap(parse_npy_vec_str)

Downloads link:

Skelibras download part 1

Skelibras download part 2

Skelibras download part 3

Skelibras download part 4

Skelibras download part 5

Cite us

About

Skellibras dataset repository

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published