2 datasets found

Groups: Referring Expression Understanding Organizations: No Organization Formats: JSON

Filter Results

  • G-Ref

    G-Ref is a dataset for referring image segmentation, comprising 104K referring language expressions for around 55K objects in about 27K images.

  • RefCOCO

    The dataset used in the paper is a benchmark for referring expression grounding, containing 142,210 referring expressions for 50,000 referents in 19,994 images.