news

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

it home reported on september 9 that getty images, the world's largest commercial photo library known for providing news, sports and entertainment photo authorization, announced last week the launch of a high-quality photo sample training data set for developers to develop and train ai models.

it home learned that the sample dataset contains 3,750 photos, divided into 15 categories such as business, education, healthcare, sports and fitness, goods and objects, illustrations, icons, etc. the dataset provides high-quality visual content on a variety of topics and can be used to train machine learning and ai models and develop a variety of application scenarios.

▲ related picture collection

currently, this dataset is available on hugging face (click here to access), but users need to agree to the terms of service and provide contact information to access the dataset for free.

getty hopes to attract businesses and developers to use its paid licensing services through this free sample data set. getty images claims to have more than 572 million photos, of which more than 200 million can be used for commercial purposes. each photo is accompanied by structured metadata, including information such as age and gender. each picture has an average of 50 keywords, which can ensure that users can safely train models and avoid the risk of infringement lawsuits.