SIREN: Semantic, Initialization-Free Registration of Multi-Robot Gaussian Splatting Maps

Olao Shorinwa, Jiankai Sun, Mac Schwager, Anirudha Majumdar
Proceedings of The 9th Conference on Robot Learning, PMLR 305:3804-3827, 2025.

Abstract

We present SIREN for registration of multi-robot Gaussian Splatting (GSplat) maps, with zero access to camera poses, images, and inter-map transforms for initialization or fusion of local submaps. To realize these capabilities, SIREN harnesses the versatility and robustness of semantics in three critical ways to derive a rigorous registration pipeline for multi-robot GSplat maps. First, SIREN utilizes semantics to identify feature-rich regions of the local maps where the registration problem is better posed, eliminating the need for any initialization which is generally required in prior work. Second, SIREN identifies candidate correspondences between Gaussians in the local maps using robust semantic features, constituting the foundation for robust geometric optimization, coarsely aligning 3D Gaussian primitives extracted from the local maps. Third, this key step enables subsequent photometric refinement of the transformation between the submaps, where SIREN leverages novel-view synthesis in GSplat maps along with a semantics-based image filter to compute a high-accuracy non-rigid transformation for the generation of a high-fidelity fused map. We demonstrate the superior performance of SIREN compared to competing baselines across a range of real-world datasets, and in particular, across the most widely used robot hardware platforms, including a manipulator, drone, and quadruped. In fact, in the most challenging scenes where accurate feature matching is extremely challenging, SIREN achieves about 90x smaller rotation errors, 300x smaller translation errors, and 44x smaller scale errors, compared to competing methods. We will release the code and provide a link to the project page after the review process.

Cite this Paper

BibTeX
@InProceedings{pmlr-v305-shorinwa25a, title = {SIREN: Semantic, Initialization-Free Registration of Multi-Robot Gaussian Splatting Maps}, author = {Shorinwa, Olao and Sun, Jiankai and Schwager, Mac and Majumdar, Anirudha}, booktitle = {Proceedings of The 9th Conference on Robot Learning}, pages = {3804--3827}, year = {2025}, editor = {Lim, Joseph and Song, Shuran and Park, Hae-Won}, volume = {305}, series = {Proceedings of Machine Learning Research}, month = {27--30 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v305/main/assets/shorinwa25a/shorinwa25a.pdf}, url = {https://proceedings.mlr.press/v305/shorinwa25a.html}, abstract = {We present SIREN for registration of multi-robot Gaussian Splatting (GSplat) maps, with zero access to camera poses, images, and inter-map transforms for initialization or fusion of local submaps. To realize these capabilities, SIREN harnesses the versatility and robustness of semantics in three critical ways to derive a rigorous registration pipeline for multi-robot GSplat maps. First, SIREN utilizes semantics to identify feature-rich regions of the local maps where the registration problem is better posed, eliminating the need for any initialization which is generally required in prior work. Second, SIREN identifies candidate correspondences between Gaussians in the local maps using robust semantic features, constituting the foundation for robust geometric optimization, coarsely aligning 3D Gaussian primitives extracted from the local maps. Third, this key step enables subsequent photometric refinement of the transformation between the submaps, where SIREN leverages novel-view synthesis in GSplat maps along with a semantics-based image filter to compute a high-accuracy non-rigid transformation for the generation of a high-fidelity fused map. We demonstrate the superior performance of SIREN compared to competing baselines across a range of real-world datasets, and in particular, across the most widely used robot hardware platforms, including a manipulator, drone, and quadruped. In fact, in the most challenging scenes where accurate feature matching is extremely challenging, SIREN achieves about 90x smaller rotation errors, 300x smaller translation errors, and 44x smaller scale errors, compared to competing methods. We will release the code and provide a link to the project page after the review process.} }
Endnote
%0 Conference Paper %T SIREN: Semantic, Initialization-Free Registration of Multi-Robot Gaussian Splatting Maps %A Olao Shorinwa %A Jiankai Sun %A Mac Schwager %A Anirudha Majumdar %B Proceedings of The 9th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Joseph Lim %E Shuran Song %E Hae-Won Park %F pmlr-v305-shorinwa25a %I PMLR %P 3804--3827 %U https://proceedings.mlr.press/v305/shorinwa25a.html %V 305 %X We present SIREN for registration of multi-robot Gaussian Splatting (GSplat) maps, with zero access to camera poses, images, and inter-map transforms for initialization or fusion of local submaps. To realize these capabilities, SIREN harnesses the versatility and robustness of semantics in three critical ways to derive a rigorous registration pipeline for multi-robot GSplat maps. First, SIREN utilizes semantics to identify feature-rich regions of the local maps where the registration problem is better posed, eliminating the need for any initialization which is generally required in prior work. Second, SIREN identifies candidate correspondences between Gaussians in the local maps using robust semantic features, constituting the foundation for robust geometric optimization, coarsely aligning 3D Gaussian primitives extracted from the local maps. Third, this key step enables subsequent photometric refinement of the transformation between the submaps, where SIREN leverages novel-view synthesis in GSplat maps along with a semantics-based image filter to compute a high-accuracy non-rigid transformation for the generation of a high-fidelity fused map. We demonstrate the superior performance of SIREN compared to competing baselines across a range of real-world datasets, and in particular, across the most widely used robot hardware platforms, including a manipulator, drone, and quadruped. In fact, in the most challenging scenes where accurate feature matching is extremely challenging, SIREN achieves about 90x smaller rotation errors, 300x smaller translation errors, and 44x smaller scale errors, compared to competing methods. We will release the code and provide a link to the project page after the review process.
APA
Shorinwa, O., Sun, J., Schwager, M. & Majumdar, A.. (2025). SIREN: Semantic, Initialization-Free Registration of Multi-Robot Gaussian Splatting Maps. Proceedings of The 9th Conference on Robot Learning, in Proceedings of Machine Learning Research 305:3804-3827 Available from https://proceedings.mlr.press/v305/shorinwa25a.html.

Related Material