Publications
1.
Song, R.; Liang, Chenwei; Xia, Y.; Zimmer, W.; Cao, H.; Caesar, H.; Festag, A.; Knoll, A.
CoDa-4DGS: Dynamic Gaussian Splatting with Context and Deformation Awareness for Autonomous Driving Proceedings Article
In: 2025 International Conference on Computer Vision (ICCV), Honolulu, Hawaii, USA, 2025.
Abstract | BibTeX | Tags: Gaussian Splatting, Intelligent Transport Systems
@inproceedings{Song:ICCV:2025,
title = {CoDa-4DGS: Dynamic Gaussian Splatting with Context and Deformation Awareness for Autonomous Driving},
author = {R. Song and Chenwei Liang and Y. Xia and W. Zimmer and H. Cao and H. Caesar and A. Festag and A. Knoll},
year = {2025},
date = {2025-06-29},
urldate = {2025-06-29},
booktitle = {2025 International Conference on Computer Vision (ICCV)},
address = {Honolulu, Hawaii, USA},
abstract = {Dynamic scene rendering opens new avenues in autonomous driving by enabling closed-loop simulations with photorealistic data, which is crucial for validating end-to-end algorithms. However, the complex and highly dynamic nature of traffic environments presents significant challenges in accurately rendering these scenes. In this paper, we introduce a novel 4D Gaussian Splatting (4DGS) approach, which incorporates context and temporal deformation awareness to improve dynamic scene rendering. Specifically, we employ a 2D semantic segmentation foundation model to self-supervise the 4D semantic features of Gaussians, ensuring meaningful contextual embedding. Simultaneously, we track the temporal deformation of each Gaussian across adjacent frames. By aggregating and encoding both semantic and temporal deformation features, each Gaussian is equipped with cues for potential deformation compensation within 3D space, facilitating a more precise representation of dynamic scenes. Experimental results show that our method improves 4DGS's ability to capture fine details in dynamic scene rendering for autonomous driving and outperforms other self-supervised methods in 4D reconstruction and novel view synthesis. Furthermore, CoDa-4DGS deforms semantic features with each Gaussian, enabling broader applications.},
keywords = {Gaussian Splatting, Intelligent Transport Systems},
pubstate = {accepted},
tppubtype = {inproceedings}
}
Dynamic scene rendering opens new avenues in autonomous driving by enabling closed-loop simulations with photorealistic data, which is crucial for validating end-to-end algorithms. However, the complex and highly dynamic nature of traffic environments presents significant challenges in accurately rendering these scenes. In this paper, we introduce a novel 4D Gaussian Splatting (4DGS) approach, which incorporates context and temporal deformation awareness to improve dynamic scene rendering. Specifically, we employ a 2D semantic segmentation foundation model to self-supervise the 4D semantic features of Gaussians, ensuring meaningful contextual embedding. Simultaneously, we track the temporal deformation of each Gaussian across adjacent frames. By aggregating and encoding both semantic and temporal deformation features, each Gaussian is equipped with cues for potential deformation compensation within 3D space, facilitating a more precise representation of dynamic scenes. Experimental results show that our method improves 4DGS's ability to capture fine details in dynamic scene rendering for autonomous driving and outperforms other self-supervised methods in 4D reconstruction and novel view synthesis. Furthermore, CoDa-4DGS deforms semantic features with each Gaussian, enabling broader applications.
2.
Jagtap, A. Dinkar; Song, R.; Sadashivaiah, S. Tiptur; Festag, A.
V2X-Gaussians: Gaussian Splatting for Multi-Agent Cooperative Dynamic Scene Reconstruction Proceedings Article
In: 36th IEEE Intelligent Vehicles Symposium (IV), Cluj-Napoca, Romania, 2025.
Abstract | BibTeX | Tags: Cooperative ITS, Gaussian Splatting, Intelligent Transport Systems, V2X communication
@inproceedings{Jagtap:IV:2025,
title = {V2X-Gaussians: Gaussian Splatting for Multi-Agent Cooperative Dynamic Scene Reconstruction},
author = {A. Dinkar Jagtap and R. Song and S. Tiptur Sadashivaiah and A. Festag},
year = {2025},
date = {2025-06-26},
urldate = {2025-03-31},
booktitle = {36th IEEE Intelligent Vehicles Symposium (IV)},
address = {Cluj-Napoca, Romania},
abstract = {Recent advances in neural rendering, such as NeRF and Gaussian Splatting, have shown great potential for dynamic scene reconstruction in intelligent vehicles. However, existing methods rely on a single ego-vehicle, suffering from limited field-of-view and occlusions, leading to incomplete reconstructions. While V2X communication may provide additional information from roadside infrastructure or other vehicless, it often degrades reconstruction quality due to sparse overlapping views. In this paper, we propose V2X-Gaussians, the first framework integrating V2X communication into Gaussian Splatting. Specifically, by leveraging deformable Gaussians and an iterative V2X-aware cross-ray densification approach, we enhance infrastructure-aided neural rendering and address view sparsity in multi-agent cooperative scenarios. In addition, to support systematic evaluation, we introduce a standardized benchmark for V2X scene reconstruction. Experiments on real-world data show that our method outperforms state-of-the-art approaches by +2.09 PSNR with only 561.8 KB for periodic V2X data exchange, highlighting the benefits of incorporating roadside infrastructure into neural rendering for intelligent transportation systems. Our code and benchmark are publicly available under an open-source license.},
keywords = {Cooperative ITS, Gaussian Splatting, Intelligent Transport Systems, V2X communication},
pubstate = {published},
tppubtype = {inproceedings}
}
Recent advances in neural rendering, such as NeRF and Gaussian Splatting, have shown great potential for dynamic scene reconstruction in intelligent vehicles. However, existing methods rely on a single ego-vehicle, suffering from limited field-of-view and occlusions, leading to incomplete reconstructions. While V2X communication may provide additional information from roadside infrastructure or other vehicless, it often degrades reconstruction quality due to sparse overlapping views. In this paper, we propose V2X-Gaussians, the first framework integrating V2X communication into Gaussian Splatting. Specifically, by leveraging deformable Gaussians and an iterative V2X-aware cross-ray densification approach, we enhance infrastructure-aided neural rendering and address view sparsity in multi-agent cooperative scenarios. In addition, to support systematic evaluation, we introduce a standardized benchmark for V2X scene reconstruction. Experiments on real-world data show that our method outperforms state-of-the-art approaches by +2.09 PSNR with only 561.8 KB for periodic V2X data exchange, highlighting the benefits of incorporating roadside infrastructure into neural rendering for intelligent transportation systems. Our code and benchmark are publicly available under an open-source license.