LanguageRefer: Spatial-Language Model for 3D Visual Grounding

Keyphrases

• Language Model 100%
• Utterance 100%
• 3D Scene 100%
• Spatial Language 100%
• 3D Visual Grounding 100%
• Task Performance 50%
• Shared Objects 50%
• Point Cloud 50%
• Robotic Applications 50%
• Spatial Reasoning 50%
• Reasoning Tasks 50%
• Learning Model 50%
• View-specific 50%
• Bounding Box 50%
• Cross-lingual Word Embeddings 50%
• Visio 50%
• Meaningful Tasks 50%
• 3D Bounding Box 50%
• Transformer Architecture 50%
• Referential Language 50%
• Grounding Problem 50%
• DistilBERT 50%
• Object Candidates 50%
• Spatial Embedding 50%

Computer Science

• 3d Scenes 100%
• Language Modeling 100%
• Robot 50%
• Annotation 50%
• Task Performance 50%
• Reasoning Task 50%
• Potential Candidate 50%
• Point Cloud 50%
• Robotics Application 50%
• DistilBERT 50%
• Candidate Object 50%