3DSRBench: A Comprehensive 3D Spatial Reasoning Benchmark

TL;DR

This paper introduces 3DSRBench, a comprehensive benchmark with 2,772 questions to evaluate and analyze the 3D spatial reasoning capabilities of large multi-modal models across diverse viewpoints and question types.

Contribution

It presents the first extensive 3D spatial reasoning benchmark, including a novel evaluation strategy and viewpoint analysis, to assess LMMs' 3D understanding.

Findings

LMMs show limitations in height, orientation, and multi-object reasoning.

Performance degrades on images with uncommon 6D viewpoints.

Benchmark provides insights for improving 3D spatial reasoning in models.

Abstract

3D spatial reasoning is the ability to analyze and interpret the positions, orientations, and spatial relationships of objects within the 3D space. This allows models to develop a comprehensive understanding of the 3D scene, enabling their applicability to a broader range of areas, such as autonomous navigation, robotics, and AR/VR. While large multi-modal models (LMMs) have achieved remarkable progress in a wide range of image and video understanding tasks, their capabilities to perform 3D spatial reasoning on diverse natural images are less studied. In this work we present the first comprehensive 3D spatial reasoning benchmark, 3DSRBench, with 2,772 manually annotated visual question-answer pairs across 12 question types. We conduct robust and thorough evaluation of 3D spatial reasoning abilities by balancing data distribution and adopting a novel FlipEval strategy. To further study the robustness of 3D spatial reasoning w.r.t. camera 3D viewpoints, our 3DSRBench includes two subsets with 3D spatial reasoning questions on paired images with common and uncommon viewpoints. We benchmark a wide range of open-sourced and proprietary LMMs, uncovering their limitations in various aspects of 3D awareness, such as height, orientation, location, and multi-object reasoning, as well as their degraded performance on images from uncommon 6D viewpoints. Our 3DSRBench provide valuable findings and insights about future development of LMMs with strong spatial reasoning abilities. Our project page is available at https://3dsrbench.github.io/.