TriCLIP-3D: A Unified Parameter-Efficient Framework for Tri-Modal 3D Visual Grounding based on CLIP

TL;DR

TriCLIP-3D introduces a unified, parameter-efficient framework leveraging a 2D CLIP model with adapter fine-tuning to enhance tri-modal 3D visual grounding, reducing complexity and improving performance.

Contribution

The paper presents a novel unified 2D pre-trained multi-modal network for processing RGB, text, and point clouds, simplifying architecture and boosting efficiency in 3D visual grounding.

Findings

Reduces trainable parameters by approximately 58%

Achieves 6.52% improvement in 3D detection

Achieves 6.25% improvement in 3D visual grounding

Abstract

3D visual grounding allows an embodied agent to understand visual information in real-world 3D environments based on human instructions, which is crucial for embodied intelligence. Existing 3D visual grounding methods typically rely on separate encoders for different modalities (e.g., RGB images, text, and 3D point clouds), resulting in large and complex models that are inefficient to train. While some approaches use pre-trained 2D multi-modal models like CLIP for 3D tasks, they still struggle with aligning point cloud data to 2D encoders. As a result, these methods continue to depend on 3D encoders for feature extraction, further increasing model complexity and training inefficiency. In this paper, we propose a unified 2D pre-trained multi-modal network to process all three modalities (RGB images, text, and point clouds), significantly simplifying the architecture. By leveraging a 2D CLIP bi-modal model with adapter-based fine-tuning, this framework effectively adapts to the tri-modal setting, improving both adaptability and performance across modalities. Our Geometric-Aware 2D-3D Feature Recovery and Fusion (GARF) module is designed to fuse geometric multi-scale features from point clouds and images. We then integrate textual features for final modality fusion and introduce a multi-modal decoder to facilitate deep cross-modal understanding. Together, our method achieves unified feature extraction and fusion across the three modalities, enabling an end-to-end 3D visual grounding model. Compared to the baseline, our method reduces the number of trainable parameters by approximately 58\%, while achieving a 6.52\% improvement in the 3D detection task and a 6.25\% improvement in the 3D visual grounding task.