Z3D: Zero-Shot 3D Visual Grounding from Images
Nikita Drozdov, Andrey Lemeshko Higher School of Economics, Higher School of Economics, Nikita Gavrilov Moscow State University, Lomonosov Moscow State University, Anton Konushin Lomonosov Moscow State University, Danila Rukhovich Institute of Mechanics and University of Luxembourg, Maksim Kolodiazhnyi Moscow State University, Lomonosov Moscow State University
原文摘要
3D visual grounding (3DVG) aims to localize objects in a 3D scene based on natural language queries. In this work, we explore zero-shot 3DVG from multi-view images alone, without requiring any geometric supervision or object priors. We introduce Z3D, a universal grounding pipeline that flexibly operates on multi-view images while optionally incorporating camera poses and depth maps. We identify key bottlenecks in prior zero-shot methods causing significant performance degradation and address them with (i) a state-of-the-art zero-shot 3D instance segmentation method to generate high-quality 3D bounding box proposals and (ii) advanced reasoning via prompt-based segmentation, which utilizes full capabilities of modern VLMs. Extensive experiments on the ScanRefer and Nr3D benchmarks demonstrate that our approach achieves state-of-the-art performance among zero-shot methods.
cs.CL
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) · 2026 · 正式发表