IR3D-Bench: Evaluating Vision-Language Model Scene Understanding as Agentic Inverse Rendering

Hengyu Liu, Chenxin Li, Zhengxin Li, Yipeng Wu, Wuyang Li, Zhiqin Yang, Zhenyuan Zhang, Yunlong Lin, Sirui Han, Brandon Y. Feng

Keywords: Vision Language Models; Inverse Rendering; Scene Understanding;

TL;DR: We introduce IR3D-Bench, a benchmark that challenges vision-language models to demonstrate real scene understanding by recreating 3D structures from images using tools, not just describing them.

Abstract: Vision-language models (VLMs) excel at descriptive tasks, but whether they truly understand scenes from visual observations remains uncertain. We introduce IR3D-Bench, a benchmark challenging VLMs to demonstrate understanding through active creation rather than passive recognition. Grounded in the analysis-by-synthesis paradigm, IR3D-Bench tasks Vision-Language Agents (VLAs) with actively using programming and rendering tools to recreate the underlying 3D structure of an input image, achieving agentic inverse rendering through tool use. This ''understanding-by-creating'' approach probes the tool-using generative capacity of VLAs, moving beyond the descriptive or conversational capacity measured by traditional scene understanding benchmarks. We provide a comprehensive suite of metrics to evaluate geometric accuracy, spatial relations, appearance attributes, and overall plausibility. Initial experiments on agentic inverse rendering powered by various state-of-the-art VLMs highlight current limitations, particularly in visual precision rather than basic tool usage. IR3D-Bench, including data and evaluation protocols, is released to facilitate systematic study and development of tool-using VLAs towards genuine scene understanding by creating.

Dataset URL: https://huggingface.co/datasets/Piang/IR3D-bench

Code URL: https://anonymous.4open.science/r/IR3D-bench-8EB2

Primary Area: Datasets & Benchmarks for applications in language modeling and vision language modeling

Submission Number: 226