Publications
1.
Jagtap, A. Dinkar; Sadashivaiah, S. Tiptur; Song, R.; Festag, A.
Hyper-V2X: Hypernetworks for Estimating Epistemic and Aleatoric Uncertainty in Cooperative Bird’s-Eye-View Semantic Segmentation Proceedings Article
In: 37th IEEE Intelligent Vehicles Symposium (IV), Detroit, USA, 2026.
Abstract | Links | BibTeX | Tags: perception
@inproceedings{Jagtap:IV:2026,
title = {Hyper-V2X: Hypernetworks for Estimating Epistemic and Aleatoric Uncertainty in Cooperative Bird’s-Eye-View Semantic Segmentation},
author = {A. Dinkar Jagtap and S. Tiptur Sadashivaiah and R. Song and A. Festag},
url = {https://arxiv.org/abs/2605.21309},
doi = {10.48550/arXiv.2605.21309},
year = {2026},
date = {2026-01-16},
urldate = {2026-05-22},
booktitle = {37th IEEE Intelligent Vehicles Symposium (IV)},
address = {Detroit, USA},
abstract = {Cooperative perception enabled by Vehicle-to-Everything (V2X) communication enhances autonomous driving safety by creating a unified environmental representation through shared sensory data. While recent works have advanced multi-agent fusion for improved perception, uncertainty quantification in such cooperative frameworks remains largely unexplored. This paper introduces Hyper-V2X, a hypernetwork-based framework for estimating both epistemic and aleatoric uncertainties in V2X-based perception. Specifically, we propose a partial weight generation scheme and V2X context embedding module that conditions a Bayesian hypernetwork on fused multi-agent features to generate weight distributions for stochastic Bird’s-Eye-View (BEV) segmentation. Unlike existing deterministic BEV models, Hyper-V2X enables efficient uncertainty estimation with little computation overhead. Our approach is architecture-agnostic, and can be seamlessly integrating with modern cooperative backbones such as CoBEVT. Experiments on the OPV2V benchmark demonstrate that Hyper-V2X provides accurate, well-calibrated uncertainty estimates and improves overall perception reliability. Source code will be made publicly available upon publication.},
keywords = {perception},
pubstate = {accepted},
tppubtype = {inproceedings}
}
Cooperative perception enabled by Vehicle-to-Everything (V2X) communication enhances autonomous driving safety by creating a unified environmental representation through shared sensory data. While recent works have advanced multi-agent fusion for improved perception, uncertainty quantification in such cooperative frameworks remains largely unexplored. This paper introduces Hyper-V2X, a hypernetwork-based framework for estimating both epistemic and aleatoric uncertainties in V2X-based perception. Specifically, we propose a partial weight generation scheme and V2X context embedding module that conditions a Bayesian hypernetwork on fused multi-agent features to generate weight distributions for stochastic Bird’s-Eye-View (BEV) segmentation. Unlike existing deterministic BEV models, Hyper-V2X enables efficient uncertainty estimation with little computation overhead. Our approach is architecture-agnostic, and can be seamlessly integrating with modern cooperative backbones such as CoBEVT. Experiments on the OPV2V benchmark demonstrate that Hyper-V2X provides accurate, well-calibrated uncertainty estimates and improves overall perception reliability. Source code will be made publicly available upon publication.
2.
Song, R.; Liang, C.; Cao, H.; Yan, Z.; Zimmer, W.; Gross, M.; Festag, A.; Knoll, A.
Collaborative Semantic Occupancy Prediction with Hybrid Feature Fusion in Connected Automated Vehicles Proceedings Article
In: 2024 Conference on Computer Vision and Pattern Recognition (CVPR), pp. 17996-18006, Seattle, WA, USA, 2024.
Abstract | Links | BibTeX | Tags: Intelligent Transport Systems, perception, sensor data sharing
@inproceedings{Song:CVPR:2024,
title = {Collaborative Semantic Occupancy Prediction with Hybrid Feature Fusion in Connected Automated Vehicles},
author = {R. Song and C. Liang and H. Cao and Z. Yan and W. Zimmer and M. Gross and A. Festag and A. Knoll},
url = {https://cvpr.thecvf.com/Conferences/2024},
doi = {10.1109/CVPR52733.2024.01704},
year = {2024},
date = {2024-02-27},
urldate = {2024-02-27},
booktitle = {2024 Conference on Computer Vision and Pattern Recognition (CVPR)},
pages = {17996-18006},
address = {Seattle, WA, USA},
abstract = {Collaborative perception in automated vehicles leverages the exchange of information between agents, aiming to elevate perception results. Previous camera-based collaborative 3D perception methods typically employ 3D bounding boxes or bird’s eye views as representations of the environment. However, these approaches fall short in offering a comprehensive 3D environmental prediction. To bridge this gap, we introduce the first method for collaborative 3D semantic occupancy prediction. Particularly, it improves local 3D semantic occupancy predictions by hybrid fusion of (i) semantic and occupancy task features, and (ii) compressed orthogonal attention features shared between vehicles. Additionally, due to the lack of a collaborative perception dataset designed for semantic occupancy prediction, we augment a current collaborative perception dataset to include 3D collaborative semantic occupancy labels for amore robust evaluation. The experimental findings highlight that: (i) our collaborative semantic occupancy predictions excel above the results from single vehicles by over 30%, and (ii) models anchored on semantic occupancy outpace state-of-the-art collaborative 3D detection techniques in subsequent perception applications, showcasing enhanced accuracy and enriched semantic-awareness in road environments.},
keywords = {Intelligent Transport Systems, perception, sensor data sharing},
pubstate = {published},
tppubtype = {inproceedings}
}
Collaborative perception in automated vehicles leverages the exchange of information between agents, aiming to elevate perception results. Previous camera-based collaborative 3D perception methods typically employ 3D bounding boxes or bird’s eye views as representations of the environment. However, these approaches fall short in offering a comprehensive 3D environmental prediction. To bridge this gap, we introduce the first method for collaborative 3D semantic occupancy prediction. Particularly, it improves local 3D semantic occupancy predictions by hybrid fusion of (i) semantic and occupancy task features, and (ii) compressed orthogonal attention features shared between vehicles. Additionally, due to the lack of a collaborative perception dataset designed for semantic occupancy prediction, we augment a current collaborative perception dataset to include 3D collaborative semantic occupancy labels for amore robust evaluation. The experimental findings highlight that: (i) our collaborative semantic occupancy predictions excel above the results from single vehicles by over 30%, and (ii) models anchored on semantic occupancy outpace state-of-the-art collaborative 3D detection techniques in subsequent perception applications, showcasing enhanced accuracy and enriched semantic-awareness in road environments.