This is the how to use page, that will help you understand how to use images.cv labeled image datasets.
Once you've downloaded and unzipped the dataset, you'll find a 'data' folder in the main directory.
Inside the 'data' folder, there are three subfolders: train, test, and val. Each subfolder contains additional folders corresponding to the categories you selected during download. The names of these folders are derived from the image annotations, separated by spaces.
The labels for the dataset are represented by the names of the folders within each subfolder (train, test, and val).
Each folder name corresponds to a specific category or class of images.
These folder names act as the labels, indicating the annotation or classification for the images contained inside.
The folder structure allows for easy organization and retrieval based on the assigned labels.
The format of all the images is .jpg. We use a compression algorithm for providing you with a well detaild but small size images
If you select the 'Augmentation options' on the download page, your training set will be randomly increased by 30% based on the options you've chosen. To identify the augmented images, navigate to the /train folder in the zip file and look for files with the 'aug' suffix.
Need more help?
Contact us
