Let’s Go Bananas: Beyond Bounding Box Representations for Fisheye Camera-Based Object Detection in Autonomous Driving
Object detection is a mature problem in autonomous driving, with pedestrian detection being one of the first commercially deployed algorithms. It has been extensively studied in the literature. However, object detection is relatively less explored for fisheye cameras used for surround-view near-fiel...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/12/3735 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839652733150298112 |
|---|---|
| author | Senthil Yogamani Ganesh Sistu Patrick Denny Jane Courtney |
| author_facet | Senthil Yogamani Ganesh Sistu Patrick Denny Jane Courtney |
| author_sort | Senthil Yogamani |
| collection | DOAJ |
| description | Object detection is a mature problem in autonomous driving, with pedestrian detection being one of the first commercially deployed algorithms. It has been extensively studied in the literature. However, object detection is relatively less explored for fisheye cameras used for surround-view near-field sensing. The standard bounding-box representation fails in fisheye cameras due to heavy radial distortion, particularly in the periphery. In this paper, a generic object detection framework is implemented using the base YOLO (You Only Look Once) detector to systematically explore various object representations using the public WoodScape dataset. First, we implement basic representations, namely the standard bounding box, the oriented bounding box, and the ellipse. Secondly, we implement a generic polygon and propose a novel curvature-adaptive polygon, which obtains an improvement of 3 mAP (mean average precision) points. A polygon is expensive to annotate and complex to use in downstream tasks; thus, it is not practical to use it in real-world applications. However, we utilize it to demonstrate that the accuracy gap between the polygon and the bounding box representation is very high due to strong distortion in fisheye cameras. This motivates the design of a distortion-aware optimal representation of the bounding box for fisheye images, which tend to be banana-shaped near the periphery. We derive a novel representation called a curved box and improve it further by leveraging vanishing-point constraints. The proposed curved box representations outperform the bounding box by 3 mAP points and the oriented bounding box by 1.6 mAP points. In addition, the camera geometry tensor is formulated to provide adaptation to non-linear fisheye camera distortion characteristics and improves the performance further by 1.4 mAP points. |
| format | Article |
| id | doaj-art-b48c4b9e441f4a23bc8f39e553232d2d |
| institution | Matheson Library |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-b48c4b9e441f4a23bc8f39e553232d2d2025-06-25T14:25:37ZengMDPI AGSensors1424-82202025-06-012512373510.3390/s25123735Let’s Go Bananas: Beyond Bounding Box Representations for Fisheye Camera-Based Object Detection in Autonomous DrivingSenthil Yogamani0Ganesh Sistu1Patrick Denny2Jane Courtney3School of Electrical & Electronic Engineering, Technological University Dublin, D07 ADY7 Dublin, IrelandD2ICE Research Centre, University of Limerick, V94 T9PX Limerick, IrelandD2ICE Research Centre, University of Limerick, V94 T9PX Limerick, IrelandSchool of Electrical & Electronic Engineering, Technological University Dublin, D07 ADY7 Dublin, IrelandObject detection is a mature problem in autonomous driving, with pedestrian detection being one of the first commercially deployed algorithms. It has been extensively studied in the literature. However, object detection is relatively less explored for fisheye cameras used for surround-view near-field sensing. The standard bounding-box representation fails in fisheye cameras due to heavy radial distortion, particularly in the periphery. In this paper, a generic object detection framework is implemented using the base YOLO (You Only Look Once) detector to systematically explore various object representations using the public WoodScape dataset. First, we implement basic representations, namely the standard bounding box, the oriented bounding box, and the ellipse. Secondly, we implement a generic polygon and propose a novel curvature-adaptive polygon, which obtains an improvement of 3 mAP (mean average precision) points. A polygon is expensive to annotate and complex to use in downstream tasks; thus, it is not practical to use it in real-world applications. However, we utilize it to demonstrate that the accuracy gap between the polygon and the bounding box representation is very high due to strong distortion in fisheye cameras. This motivates the design of a distortion-aware optimal representation of the bounding box for fisheye images, which tend to be banana-shaped near the periphery. We derive a novel representation called a curved box and improve it further by leveraging vanishing-point constraints. The proposed curved box representations outperform the bounding box by 3 mAP points and the oriented bounding box by 1.6 mAP points. In addition, the camera geometry tensor is formulated to provide adaptation to non-linear fisheye camera distortion characteristics and improves the performance further by 1.4 mAP points.https://www.mdpi.com/1424-8220/25/12/3735automated drivingobject detectionsurround view camerasfisheye cameras |
| spellingShingle | Senthil Yogamani Ganesh Sistu Patrick Denny Jane Courtney Let’s Go Bananas: Beyond Bounding Box Representations for Fisheye Camera-Based Object Detection in Autonomous Driving Sensors automated driving object detection surround view cameras fisheye cameras |
| title | Let’s Go Bananas: Beyond Bounding Box Representations for Fisheye Camera-Based Object Detection in Autonomous Driving |
| title_full | Let’s Go Bananas: Beyond Bounding Box Representations for Fisheye Camera-Based Object Detection in Autonomous Driving |
| title_fullStr | Let’s Go Bananas: Beyond Bounding Box Representations for Fisheye Camera-Based Object Detection in Autonomous Driving |
| title_full_unstemmed | Let’s Go Bananas: Beyond Bounding Box Representations for Fisheye Camera-Based Object Detection in Autonomous Driving |
| title_short | Let’s Go Bananas: Beyond Bounding Box Representations for Fisheye Camera-Based Object Detection in Autonomous Driving |
| title_sort | let s go bananas beyond bounding box representations for fisheye camera based object detection in autonomous driving |
| topic | automated driving object detection surround view cameras fisheye cameras |
| url | https://www.mdpi.com/1424-8220/25/12/3735 |
| work_keys_str_mv | AT senthilyogamani letsgobananasbeyondboundingboxrepresentationsforfisheyecamerabasedobjectdetectioninautonomousdriving AT ganeshsistu letsgobananasbeyondboundingboxrepresentationsforfisheyecamerabasedobjectdetectioninautonomousdriving AT patrickdenny letsgobananasbeyondboundingboxrepresentationsforfisheyecamerabasedobjectdetectioninautonomousdriving AT janecourtney letsgobananasbeyondboundingboxrepresentationsforfisheyecamerabasedobjectdetectioninautonomousdriving |