E3VS-Bench: A Benchmark for Viewpoint-Dependent Active Perception in 3D Gaussian Splatting Scenes

TL;DR

E3VS-Bench is a new benchmark for evaluating embodied agents' ability to perform viewpoint-dependent visual search in 3D scenes, emphasizing the importance of active perception with 5-DoF control.

Contribution

The paper introduces E3VS-Bench, a high-fidelity 3D scene benchmark with questions requiring active viewpoint control, enabling evaluation of perception and planning in unrestricted 5-DoF environments.

Findings

Models perform significantly worse than humans on viewpoint-dependent tasks.

Photorealistic rendering with 3D Gaussian Splatting captures fine details for complex questions.

Current models show limitations in active perception and viewpoint planning.

Abstract

Visual search in 3D environments requires embodied agents to actively explore their surroundings and acquire task-relevant evidence. However, existing visual search and embodied AI benchmarks, including EQA, typically rely on static observations or constrained egocentric motion, and thus do not explicitly evaluate fine-grained viewpoint-dependent phenomena that arise under unrestricted 5-DoF viewpoint control in real-world 3D environments, such as visibility changes caused by vertical viewpoint shifts, revealing contents inside containers, and disambiguating object attributes that are only observable from specific angles. To address this limitation, we introduce {E3VS-Bench}, a benchmark for embodied 3D visual search where agents must control their viewpoints in 5-DoF to gather viewpoint-dependent evidence for question answering. E3VS-Bench consists of 99 high-fidelity 3D scenes reconstructed using 3D Gaussian Splatting and 2,014 question-driven episodes. 3D Gaussian Splatting enables photorealistic free-viewpoint rendering that preserves fine-grained visual details (e.g., small text and subtle attributes) often degraded in mesh-based simulators, thereby allowing the construction of questions that cannot be answered from a single view and instead require active inspection across viewpoints in 5-DoF. We evaluate multiple state-of-the-art VLMs and compare their performance with humans. Despite strong 2D reasoning ability, all models exhibit a substantial gap from humans, highlighting limitations in active perception and coherent viewpoint planning specifically under full 5-DoF viewpoint changes.