VLM-Grounder: A VLM Agent for Zero-Shot 3D Visual Grounding

Runsen Xu, Zhiwei Huang, Tai Wang, Yilun Chen, Jiangmiao Pang, Dahua Lin
Proceedings of The 8th Conference on Robot Learning, PMLR 270:3961-3985, 2025.

Abstract

3D visual grounding is crucial for robots, requiring integration of natural language and 3D scene understanding. Traditional methods depend on supervised learning with 3D point clouds are limited by scarce datasets. Recently zero-shot methods leveraging LLMs have been proposed to address the data issue. While effective, these methods often miss detailed scene context, limiting their ability to handle complex queries. In this work, we present VLM-Grounder, a novel framework using vision-language models (VLMs) for zero-shot 3D visual grounding based solely on 2D images. VLM-Grounder dynamically stitches image sequences, employs a grounding and feedback scheme to find the target object, and uses a multi-view ensemble projection to accurately estimate 3D bounding boxes. Experiments on ScanRefer and Nr3D datasets show VLM-Grounder outperforms previous zero-shot methods, achieving 51.6% Acc@0.25 on ScanRefer and 48.0% Acc on Nr3D, without relying on 3D geometry or object priors.

Cite this Paper

BibTeX
@InProceedings{pmlr-v270-xu25c, title = {VLM-Grounder: A VLM Agent for Zero-Shot 3D Visual Grounding}, author = {Xu, Runsen and Huang, Zhiwei and Wang, Tai and Chen, Yilun and Pang, Jiangmiao and Lin, Dahua}, booktitle = {Proceedings of The 8th Conference on Robot Learning}, pages = {3961--3985}, year = {2025}, editor = {Agrawal, Pulkit and Kroemer, Oliver and Burgard, Wolfram}, volume = {270}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v270/main/assets/xu25c/xu25c.pdf}, url = {https://proceedings.mlr.press/v270/xu25c.html}, abstract = {3D visual grounding is crucial for robots, requiring integration of natural language and 3D scene understanding. Traditional methods depend on supervised learning with 3D point clouds are limited by scarce datasets. Recently zero-shot methods leveraging LLMs have been proposed to address the data issue. While effective, these methods often miss detailed scene context, limiting their ability to handle complex queries. In this work, we present VLM-Grounder, a novel framework using vision-language models (VLMs) for zero-shot 3D visual grounding based solely on 2D images. VLM-Grounder dynamically stitches image sequences, employs a grounding and feedback scheme to find the target object, and uses a multi-view ensemble projection to accurately estimate 3D bounding boxes. Experiments on ScanRefer and Nr3D datasets show VLM-Grounder outperforms previous zero-shot methods, achieving 51.6% Acc@0.25 on ScanRefer and 48.0% Acc on Nr3D, without relying on 3D geometry or object priors.} }
Endnote
%0 Conference Paper %T VLM-Grounder: A VLM Agent for Zero-Shot 3D Visual Grounding %A Runsen Xu %A Zhiwei Huang %A Tai Wang %A Yilun Chen %A Jiangmiao Pang %A Dahua Lin %B Proceedings of The 8th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Pulkit Agrawal %E Oliver Kroemer %E Wolfram Burgard %F pmlr-v270-xu25c %I PMLR %P 3961--3985 %U https://proceedings.mlr.press/v270/xu25c.html %V 270 %X 3D visual grounding is crucial for robots, requiring integration of natural language and 3D scene understanding. Traditional methods depend on supervised learning with 3D point clouds are limited by scarce datasets. Recently zero-shot methods leveraging LLMs have been proposed to address the data issue. While effective, these methods often miss detailed scene context, limiting their ability to handle complex queries. In this work, we present VLM-Grounder, a novel framework using vision-language models (VLMs) for zero-shot 3D visual grounding based solely on 2D images. VLM-Grounder dynamically stitches image sequences, employs a grounding and feedback scheme to find the target object, and uses a multi-view ensemble projection to accurately estimate 3D bounding boxes. Experiments on ScanRefer and Nr3D datasets show VLM-Grounder outperforms previous zero-shot methods, achieving 51.6% Acc@0.25 on ScanRefer and 48.0% Acc on Nr3D, without relying on 3D geometry or object priors.
APA
Xu, R., Huang, Z., Wang, T., Chen, Y., Pang, J. & Lin, D.. (2025). VLM-Grounder: A VLM Agent for Zero-Shot 3D Visual Grounding. Proceedings of The 8th Conference on Robot Learning, in Proceedings of Machine Learning Research 270:3961-3985 Available from https://proceedings.mlr.press/v270/xu25c.html.

Related Material