LocateAnything Open-Sources Multimodal Dataset with 12M Images

Zhiding Yu's team open-sourced LocateAnything-Data, a massive spatial dataset with 12 million images and 785 million annotations, to help the community build advanced multimodal models.

2026-07-27 ~ 2026-07-28 · 2 related posts

• LocateAnything-Data opens up 12M images and 785M spatial annotations for multimodal training — ZhidingYu · 2026-07-27

1 near-duplicate retellings: ZhidingYu