3 datasets found

Groups: Image Segmentation

Filter Results

  • 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.

  • RefCOCO, RefCOCO+, and RefCOCOg

    Visual Grounding is a task that aims to locate a target object according to a natural language expression. The dataset used in this paper is RefCOCO, RefCOCO+, and RefCOCOg.

  • Visual Genome

    The Visual Genome dataset is a large-scale visual question answering dataset, containing 1.5 million images, each with 15-30 annotated entities, attributes, and relationships.