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.
@article{arxiv.2506.23329,
title = {IR3D-Bench: Evaluating Vision-Language Model Scene Understanding as Agentic Inverse Rendering},
author = {Parker Liu and Chenxin Li and Zhengxin Li and Yipeng Wu and Wuyang Li and Zhiqin Yang and Zhenyuan Zhang and Yunlong Lin and Sirui Han and Brandon Y. Feng},
journal= {arXiv preprint arXiv:2506.23329},
year = {2025}
}