N3D-VLM: Native 3D Grounding Enables Accurate Spatial Reasoning in Vision-Language Models

TL;DR

N3D-VLM introduces a unified framework that integrates native 3D object perception with visual reasoning, significantly enhancing spatial understanding and grounding accuracy in vision-language models.

Contribution

The paper presents a novel approach combining native 3D perception with spatial reasoning, supported by a large-scale 3D dataset generated through depth-based annotation lifting.

Findings

Achieves state-of-the-art results on 3D grounding tasks.

Surpasses existing methods in 3D spatial reasoning.

Uses a scalable data pipeline for diverse 3D annotations.

Abstract

While current multimodal models can answer questions based on 2D images, they lack intrinsic 3D object perception, limiting their ability to comprehend spatial relationships and depth cues in 3D scenes. In this work, we propose N3D-VLM, a novel unified framework that seamlessly integrates native 3D object perception with 3D-aware visual reasoning, enabling both precise 3D grounding and interpretable spatial understanding. Unlike conventional end-to-end models that directly predict answers from RGB/RGB-D inputs, our approach equips the model with native 3D object perception capabilities, enabling it to directly localize objects in 3D space based on textual descriptions. Building upon accurate 3D object localization, the model further performs explicit reasoning in 3D, achieving more interpretable and structured spatial understanding. To support robust training for these capabilities, we develop a scalable data construction pipeline that leverages depth estimation to lift large-scale 2D annotations into 3D space, significantly increasing the diversity and coverage for 3D object grounding data, yielding over six times larger than the largest existing single-image 3D detection dataset. Moreover, the pipeline generates spatial question-answering datasets that target chain-of-thought (CoT) reasoning in 3D, facilitating joint training for both 3D object localization and 3D spatial reasoning. Experimental results demonstrate that our unified framework not only achieves state-of-the-art performance on 3D grounding tasks, but also consistently surpasses existing methods in 3D spatial reasoning in vision-language model.