Spatial language Integrating Model (SLIM)

This dataset consists of virtual scenes rendered in MuJoCo with multiple views each presented in multiple modalities: image, and synthetic or natural language descriptions. Each scene consists of two or three objects placed on a square walled room, and for each of the 10 camera viewpoint we render a 3D view of the scene as seen from that viewpoint as well as a synthetically generated description of the scene.

No items found.