3D-GRES: Generalized 3D Referring Expression Segmentation

TL;DR

This paper introduces 3D-GRES, a novel task for segmenting multiple objects in 3D space based on natural language, along with a new model MDIN and a dedicated dataset Multi3DRes, advancing multi-object 3D scene understanding.

Contribution

The paper presents the first generalized framework for multi-object 3D referring expression segmentation, including a new model and dataset to handle complex multi-object scenarios.

Findings

MDIN significantly outperforms existing models on Multi3DRes.

The proposed approach effectively segments multiple 3D objects based on natural language.

The new dataset enables comprehensive evaluation of multi-object 3D segmentation methods.

Abstract

3D Referring Expression Segmentation (3D-RES) is dedicated to segmenting a specific instance within a 3D space based on a natural language description. However, current approaches are limited to segmenting a single target, restricting the versatility of the task. To overcome this limitation, we introduce Generalized 3D Referring Expression Segmentation (3D-GRES), which extends the capability to segment any number of instances based on natural language instructions. In addressing this broader task, we propose the Multi-Query Decoupled Interaction Network (MDIN), designed to break down multi-object segmentation tasks into simpler, individual segmentations. MDIN comprises two fundamental components: Text-driven Sparse Queries (TSQ) and Multi-object Decoupling Optimization (MDO). TSQ generates sparse point cloud features distributed over key targets as the initialization for queries. Meanwhile, MDO is tasked with assigning each target in multi-object scenarios to different queries while maintaining their semantic consistency. To adapt to this new task, we build a new dataset, namely Multi3DRes. Our comprehensive evaluations on this dataset demonstrate substantial enhancements over existing models, thus charting a new path for intricate multi-object 3D scene comprehension. The benchmark and code are available at https://github.com/sosppxo/MDIN.