ScanReason: Empowering 3D Visual Grounding with Reasoning Capabilities

TL;DR

ScanReason introduces a new benchmark and approach for 3D visual grounding that incorporates reasoning capabilities to interpret implicit instructions, advancing the understanding of human intentions in 3D scenes.

Contribution

The paper presents a novel task, benchmark, and method that integrate reasoning with 3D grounding, utilizing multi-modal large language models and a chain-of-grounding mechanism.

Findings

Effective performance on the ScanReason benchmark.

Improved accuracy in reasoning-based 3D grounding.

Validation of the approach through extensive experiments.

Abstract

Although great progress has been made in 3D visual grounding, current models still rely on explicit textual descriptions for grounding and lack the ability to reason human intentions from implicit instructions. We propose a new task called 3D reasoning grounding and introduce a new benchmark ScanReason which provides over 10K question-answer-location pairs from five reasoning types that require the synerization of reasoning and grounding. We further design our approach, ReGround3D, composed of the visual-centric reasoning module empowered by Multi-modal Large Language Model (MLLM) and the 3D grounding module to obtain accurate object locations by looking back to the enhanced geometry and fine-grained details from the 3D scenes. A chain-of-grounding mechanism is proposed to further boost the performance with interleaved reasoning and grounding steps during inference. Extensive experiments on the proposed benchmark validate the effectiveness of our proposed approach.