Labl.it

Image dataset labeling utility for image classification tasks

Main script at link

usage: main.py [-h] -k KEYWORDS -c CLASSES -s SHORTCUTS [-f FILENAME]

optional arguments:

-h, --help show this help message and exit

-k KEYWORDS, --keywords KEYWORDS

                    keywords in the file path to be found, must be comma-separted if
                    
                    more than 1 Eg. class1,positve,png or
                    
                    class2,negative,jpg

-c CLASSES, --classes CLASSES

                    names of target classes to be used, must be comma-
                    
                    separted and mentioned ordinally (as we'll use 0 for
                    
                    first class, 1 for second and so on) . Eg. dog,cat or
                    
                    abnormal,normal

-s SHORTCUTS, --shortcuts SHORTCUTS

                    keyboard shortcuts to be used to trigger label
                    
                    buttons, must be comma-separted and in same order as
                    
                    the classes in -c argument. Eg. d,g (for dog,cat) or
                    
                    a,n (for abnormal,normal)

-f FILENAME, --filename FILENAME

                    filename used to save the final label files, filename
                    
                    must be given without extension (mylabel for
                    
                    mylabels.csv). if this argument is empty, default name
                    
                    is "labels.csv"