Dataset Library
A curated catalog of vision datasets across every domain — with size, task, and license at a glance.
Web-scale image-text dataset behind CLIP-style and diffusion models.
Multi-institutional MRI for 3D brain tumor segmentation challenges.
Full sensor-suite (camera, LiDAR, radar) driving dataset for 3D perception.
Dermoscopy images for melanoma detection — a standard medical imaging benchmark.
Defect/anomaly detection benchmark for industrial quality inspection.
Large aerial images with oriented bounding boxes for remote sensing.
Massive, richly annotated detection dataset with hierarchical labels.
Dense semantic segmentation across diverse scenes and stuff/thing classes.
Scene-text detection and recognition benchmarks, real and synthetic.
Large-scale human action video dataset for spatiotemporal models.
High-quality pixel annotations of street scenes for driving perception.
Richly annotated 3D CAD models for reconstruction and 3D vision.
The de-facto benchmark for object detection and instance segmentation.
A classic, compact detection/segmentation benchmark — great for learning.
Foundational autonomous-driving benchmark with LiDAR + stereo + boxes.
The benchmark that launched deep learning for vision; standard pretraining dataset.