ManipMob-MMKG

This is a repo for 'Scene-Driven Multimodal Knowledge Graph Construction for Embodied AI', and our website is https://sites.google.com/view/manipmob-mmkg

Schema Visualization

Downstream Task

Visual Language Navigation

3D Object Language Grounding