Towards Robust Autonomous Driving: Out-of-Distribution Object Detection in Bird's Eye View Space
In autonomous driving, understanding the surroundings is crucial for safety. Since most object detection systems are designed to identify known objects, they may miss unknown or novel objects, which can be dangerous. This study addresses Out-Of-Distribution (OOD) detection for vehicle-like unknown o...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Open Journal of Vehicular Technology |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11031213/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839646296514756608 |
|---|---|
| author | Muhammad Asad Ihsan Ullah Ganesh Sistu Michael G. Madden |
| author_facet | Muhammad Asad Ihsan Ullah Ganesh Sistu Michael G. Madden |
| author_sort | Muhammad Asad |
| collection | DOAJ |
| description | In autonomous driving, understanding the surroundings is crucial for safety. Since most object detection systems are designed to identify known objects, they may miss unknown or novel objects, which can be dangerous. This study addresses Out-Of-Distribution (OOD) detection for vehicle-like unknown objects within the Bird's Eye View (BeV) space, a top-down representation of the environment that provides a comprehensive spatial layout crucial for scene understanding. Enhancing the model's adaptability to unfamiliar objects, we present two novel methods for detecting unknown objects in BeV space. Specifically, we introduce random patches and OOD objects in the environment to help the model identify both known objects, such as vehicles, and OOD objects. We also introduce a new dataset, NuScenesOOD, derived from the NuScenes dataset, which augments vehicles with patterns and shapes to challenge the model. Additionally, we address challenges such as patch size inconsistency and occlusion from moving frames in BeV space. Our method targets vehicle-shaped anomalies in the planar driving space, maintaining high accuracy for known and enhancing detection of unknown objects. This research contributes to making future autonomous vehicles safer by improving their ability to detect diverse vehicle like OOD objects in their environment. |
| format | Article |
| id | doaj-art-d59d87af308c475b80ade24664b96b51 |
| institution | Matheson Library |
| issn | 2644-1330 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of Vehicular Technology |
| spelling | doaj-art-d59d87af308c475b80ade24664b96b512025-06-30T23:01:17ZengIEEEIEEE Open Journal of Vehicular Technology2644-13302025-01-0161673168510.1109/OJVT.2025.357934111031213Towards Robust Autonomous Driving: Out-of-Distribution Object Detection in Bird's Eye View SpaceMuhammad Asad0https://orcid.org/0000-0002-6398-0320Ihsan Ullah1https://orcid.org/0000-0002-7964-5199Ganesh Sistu2https://orcid.org/0009-0003-1683-9257Michael G. Madden3https://orcid.org/0000-0002-4443-7285Machine Learning Research Group, School of Computer Science and Insight Research Ireland Centre for Data Analytics, University of Galway, Galway, IrelandMachine Learning Research Group, School of Computer Science and Insight Research Ireland Centre for Data Analytics, University of Galway, Galway, IrelandValeo Vision Systems, Tuam, IrelandMachine Learning Research Group, School of Computer Science and Insight Research Ireland Centre for Data Analytics, University of Galway, Galway, IrelandIn autonomous driving, understanding the surroundings is crucial for safety. Since most object detection systems are designed to identify known objects, they may miss unknown or novel objects, which can be dangerous. This study addresses Out-Of-Distribution (OOD) detection for vehicle-like unknown objects within the Bird's Eye View (BeV) space, a top-down representation of the environment that provides a comprehensive spatial layout crucial for scene understanding. Enhancing the model's adaptability to unfamiliar objects, we present two novel methods for detecting unknown objects in BeV space. Specifically, we introduce random patches and OOD objects in the environment to help the model identify both known objects, such as vehicles, and OOD objects. We also introduce a new dataset, NuScenesOOD, derived from the NuScenes dataset, which augments vehicles with patterns and shapes to challenge the model. Additionally, we address challenges such as patch size inconsistency and occlusion from moving frames in BeV space. Our method targets vehicle-shaped anomalies in the planar driving space, maintaining high accuracy for known and enhancing detection of unknown objects. This research contributes to making future autonomous vehicles safer by improving their ability to detect diverse vehicle like OOD objects in their environment.https://ieeexplore.ieee.org/document/11031213/Autonomous drivingout-of-distribution (OOD) detectionbird's eye view (BeV)feature spaceobject detection |
| spellingShingle | Muhammad Asad Ihsan Ullah Ganesh Sistu Michael G. Madden Towards Robust Autonomous Driving: Out-of-Distribution Object Detection in Bird's Eye View Space IEEE Open Journal of Vehicular Technology Autonomous driving out-of-distribution (OOD) detection bird's eye view (BeV) feature space object detection |
| title | Towards Robust Autonomous Driving: Out-of-Distribution Object Detection in Bird's Eye View Space |
| title_full | Towards Robust Autonomous Driving: Out-of-Distribution Object Detection in Bird's Eye View Space |
| title_fullStr | Towards Robust Autonomous Driving: Out-of-Distribution Object Detection in Bird's Eye View Space |
| title_full_unstemmed | Towards Robust Autonomous Driving: Out-of-Distribution Object Detection in Bird's Eye View Space |
| title_short | Towards Robust Autonomous Driving: Out-of-Distribution Object Detection in Bird's Eye View Space |
| title_sort | towards robust autonomous driving out of distribution object detection in bird x0027 s eye view space |
| topic | Autonomous driving out-of-distribution (OOD) detection bird's eye view (BeV) feature space object detection |
| url | https://ieeexplore.ieee.org/document/11031213/ |
| work_keys_str_mv | AT muhammadasad towardsrobustautonomousdrivingoutofdistributionobjectdetectioninbirdx0027seyeviewspace AT ihsanullah towardsrobustautonomousdrivingoutofdistributionobjectdetectioninbirdx0027seyeviewspace AT ganeshsistu towardsrobustautonomousdrivingoutofdistributionobjectdetectioninbirdx0027seyeviewspace AT michaelgmadden towardsrobustautonomousdrivingoutofdistributionobjectdetectioninbirdx0027seyeviewspace |