VLM-3R: Vision-Language Models Augmented with Instruction-Aligned 3D Reconstruction

TL;DR

VLM-3R is a unified vision-language framework that enhances 3D spatial understanding from monocular videos through instruction tuning and a new benchmark, advancing human-like visual-spatial reasoning.

Contribution

The paper introduces VLM-3R, a novel model integrating 3D reconstructive instruction tuning and a new benchmark for temporal reasoning in monocular videos.

Findings

VLM-3R achieves robust 3D spatial reasoning from monocular videos.

The model effectively aligns spatial context with language instructions.

VLM-3R demonstrates superior accuracy and scalability in temporal 3D understanding.

Abstract

The rapid advancement of Large Multimodal Models (LMMs) for 2D images and videos has motivated extending these models to understand 3D scenes, aiming for human-like visual-spatial intelligence. Nevertheless, achieving deep spatial understanding comparable to human capabilities poses significant challenges in model encoding and data acquisition. Existing methods frequently depend on external depth sensors for geometry capture or utilize off-the-shelf algorithms for pre-constructing 3D maps, thereby limiting their scalability, especially with prevalent monocular video inputs and for time-sensitive applications. In this work, we introduce VLM-3R, a unified framework for Vision-Language Models (VLMs) that incorporates 3D Reconstructive instruction tuning. VLM-3R processes monocular video frames by employing a geometry encoder to derive implicit 3D tokens that represent spatial understanding. Leveraging our Spatial-Visual-View Fusion and over 200K curated 3D reconstructive instruction tuning question-answer (QA) pairs, VLM-3R effectively aligns real-world spatial context with language instructions. This enables monocular 3D spatial assistance and embodied reasoning. To facilitate the evaluation of temporal reasoning, we introduce the Vision-Spatial-Temporal Intelligence benchmark, featuring over 138.6K QA pairs across five distinct tasks focused on evolving spatial relationships. Extensive experiments demonstrate that our model, VLM-3R, not only facilitates robust visual-spatial reasoning but also enables the understanding of temporal 3D context changes, excelling in both accuracy and scalability.