LanguageRefer: Spatial-Language Model for 3D Visual Grounding

TL;DR

LanguageRefer is a transformer-based model that combines spatial and language embeddings to accurately identify objects in 3D scenes from natural language references, advancing robotic understanding of human instructions.

Contribution

It introduces a novel spatial-language model for 3D visual grounding using a transformer architecture with spatial and language embeddings, demonstrating competitive performance.

Findings

Performs well on ReferIt3D datasets

Effective spatial reasoning decoupled from perception noise

Accurate in view-dependent utterance scenarios

Abstract

For robots to understand human instructions and perform meaningful tasks in the near future, it is important to develop learned models that comprehend referential language to identify common objects in real-world 3D scenes. In this paper, we introduce a spatial-language model for a 3D visual grounding problem. Specifically, given a reconstructed 3D scene in the form of point clouds with 3D bounding boxes of potential object candidates, and a language utterance referring to a target object in the scene, our model successfully identifies the target object from a set of potential candidates. Specifically, LanguageRefer uses a transformer-based architecture that combines spatial embedding from bounding boxes with fine-tuned language embeddings from DistilBert to predict the target object. We show that it performs competitively on visio-linguistic datasets proposed by ReferIt3D. Further, we analyze its spatial reasoning task performance decoupled from perception noise, the accuracy of view-dependent utterances, and viewpoint annotations for potential robotics applications.