Hierarchical Collaborative Fusion for 3D Instance-aware Referring Expression Segmentation

TL;DR

This paper introduces HCF-RES, a multi-modal framework that enhances 3D referring expression segmentation by combining hierarchical visual semantics with progressive multi-level fusion, leading to state-of-the-art results.

Contribution

The paper proposes a novel hierarchical semantic decomposition and multi-level fusion approach to improve 3D referring expression segmentation accuracy.

Findings

Achieves state-of-the-art performance on ScanRefer and Multi3DRefer datasets.

Effectively preserves object boundaries during 2D-to-3D projection.

Enhances multi-modal feature integration through adaptive weighting.

Abstract

Generalised 3D Referring Expression Segmentation (3D-GRES) localizes objects in 3D scenes based on natural language, even when descriptions match multiple or zero targets. Existing methods rely solely on sparse point clouds, lacking rich visual semantics for fine-grained descriptions. We propose HCF-RES, a multi-modal framework with two key innovations. First, Hierarchical Visual Semantic Decomposition leverages SAM instance masks to guide CLIP encoding at dual granularities -- pixel-level and instance-level features -- preserving object boundaries during 2D-to-3D projection. Second, Progressive Multi-level Fusion integrates representations through intra-modal collaboration, cross-modal adaptive weighting between 2D semantic and 3D geometric features, and language-guided refinement. HCF-RES achieves state-of-the-art results on both ScanRefer and Multi3DRefer.