NVEagle/LocateAnything-Data · Datasets at Hugging Face

Overview Data coverage Spatial supervision Release format Annotation record Download Download selected datasets Inspect annotations Visualize a sample Read for training Map a record back to its source image Media downloaded from upstream sources Prepare an upstream media dataset License Information Copyright concerns Acknowledgements Citation LocateAnything-Data 中文 · Paper · Model · Code

Overview LocateAnything-Data is the public training-data release for LocateAnything: Fast and High-Quality Vision-Language Grounding with Parallel Box Decoding.

LocateAnything formulates detection and visual grounding as a unified vision-language task. Given an image and a category, phrase, text string, or action-oriented instruction, the model predicts the corresponding bounding boxes or points. The data spans natural images, dense scenes, people, autonomous driving, embodied interaction, graphical user interfaces, scene text, documents, and tables.

The paper's Parallel Box Decoding treats a box or point as a structured geometric unit instead of generating its coordinates independently. LocateAnything-Data provides the diverse spatial supervision used to train this unified formulation across visual domains.

Data coverage The following grouping describes the main visual domains in this release. Several datasets naturally support more than one domain or task.