Open Images

~9M images annotated with image-level labels, bounding boxes, segmentations, and visual relationships.

Papers using Open Images (3)

TriLite: Efficient Weakly Supervised Object Localization with Universal Visual Features and Tri-Region Disentanglement2026Looking Locally: Object-Centric Vision Transformers as Foundation Models for Efficient Segmentation2025Detecting Visual Relationships Using Box Attention2018