Discover published sets by community

A large dataset of handwritten digits widely used for training and testing in the field of machine learning. It has become a benchmark for evaluating classification algorithms.

An extension of the original COCO dataset, it contains more diverse images and annotations including object detection, segmentation, and captioning for a deeper understanding of image content.

Consists of around one million labeled images for each of 10 scene categories and 20 object categories. The dataset is used for various tasks in scene understanding and object recognition.

A dataset for mobile robotics and autonomous driving research, featuring a variety of image, LIDAR, and GPS/IMU data from urban environments.

Similar to CIFAR-10 but with 100 classes, each containing 600 images, for a total of 60,000 images. Each class has 500 training images and 100 test images.

A real-world image dataset obtained from Google Street View images. It contains over 600,000 digit images and is used for developing machine learning and object recognition algorithms.

A dataset for object detection and image classification that provides standardized image data sets for object class recognition. It has been a benchmark for various tasks including classification, detection, and segmentation.

A dataset with RGB and depth images recorded by the Microsoft Kinect. Ideal for tasks involving 3D scene understanding and indoor scene reconstruction.

A dataset consisting of 60,000 32x32 color images in 10 different classes, with 6000 images per class. It is commonly used to train machine learning and computer vision algorithms.

A large-scale dataset consisting of millions of labeled high-resolution images across thousands of categories. It is widely used for training and benchmarking image classification algorithms.

A dataset with rich annotations for object detection, segmentation, and captioning. It provides images with context for the objects, emphasizing the scene understanding.

A large-scale face attributes dataset containing more than 200K celebrity images, each with 40 attribute annotations. It is used for developing algorithms in facial recognition and attribute detection.

A pet dataset consisting of images of cats and dogs, with annotations for breed classification. It includes 37 categories with roughly 200 images for each class.

A collection of ~9 million images annotated with image labels, object bounding boxes, object segmentation masks, and visual relationships. It's a diverse and large-scale dataset for image understanding tasks.