Talk2Radar: Bridging Natural Language with 4D mmWave Radar for 3D Referring Expression Comprehension

TL;DR

This paper introduces Talk2Radar, a novel dataset and model for 3D referring expression comprehension using 4D mmWave radar data, enabling natural language understanding of objects in radar scenes.

Contribution

It is the first to bridge natural language with 4D radar data for 3D visual grounding, proposing a new dataset and a state-of-the-art model for this task.

Findings

Achieved SOTA performance on Talk2Radar dataset

Designed Deformable-FPN for efficient point cloud feature extraction

Developed Gated Graph Fusion for effective cross-modal feature fusion

Abstract

Embodied perception is essential for intelligent vehicles and robots in interactive environmental understanding. However, these advancements primarily focus on vision, with limited attention given to using 3D modeling sensors, restricting a comprehensive understanding of objects in response to prompts containing qualitative and quantitative queries. Recently, as a promising automotive sensor with affordable cost, 4D millimeter-wave radars provide denser point clouds than conventional radars and perceive both semantic and physical characteristics of objects, thereby enhancing the reliability of perception systems. To foster the development of natural language-driven context understanding in radar scenes for 3D visual grounding, we construct the first dataset, Talk2Radar, which bridges these two modalities for 3D Referring Expression Comprehension (REC). Talk2Radar contains 8,682 referring prompt samples with 20,558 referred objects. Moreover, we propose a novel model, T-RadarNet, for 3D REC on point clouds, achieving State-Of-The-Art (SOTA) performance on the Talk2Radar dataset compared to counterparts. Deformable-FPN and Gated Graph Fusion are meticulously designed for efficient point cloud feature modeling and cross-modal fusion between radar and text features, respectively. Comprehensive experiments provide deep insights into radar-based 3D REC. We release our project at https://github.com/GuanRunwei/Talk2Radar.