Expandable Spherical Projection and Feature Concatenation Methods for Real-Time Road Object Detection Using Fisheye Image

Abstract

Fisheye lens cameras are widely used in such applications where a large field of view (FOV) is necessary. A large FOV can provide an enhanced understanding of the surrounding environment and can be an effective solution for detecting the objects in automotive applications. However, this comes with the cost of strong radial distortions and irregular size of objects depending on the location in an image. Therefore, we propose a new fisheye image warping method called Expandable Spherical Projection to expand the center and boundary regions in which smaller objects are mostly located. The proposed method produces undistorted objects especially in the image boundary and a less unwanted background in the bounding boxes. Additionally, we propose three multi-scale feature concatenation methods and provide the analysis of the influence from the three concatenation methods in a real-time object detector. Multiple fisheye image datasets are employed to demonstrate the effectiveness of the proposed projection and feature concatenation methods. From the experimental results, we find that the proposed Expandable Spherical projection and the LCat feature concatenation yield the best AP performance, which is up to 4.7% improvement compared to the original fisheye image datasets and the baseline model.

PGDS-YOLOv8s: An Improved YOLOv8s Model for Object Detection in Fisheye Images

“…Table 3 shows a comparison of the experimental results of the proposed method with the state-of-the-art object detection methods for fisheye images. Among them, Kim et al [44] proposed three multi-scale feature connection models for detecting fisheye images, including Long-Skip Concatenation Model (LCat), Short-Skip Concatenation Model (SCat), and Short-Long-Skip Concatenation Model (SLCat). In addition, the edge continuity distortion-aware block (ECDAB) [45] and SphereConv [46] are introduced into the YOLOv8s model for comparison experiments with the proposed model, respectively.…”

An improved YOLOv8s model based on sphere convolution for object detection in fisheye images

“…The field of view of a fisheye lens can reach 180° to 270°, and its field of view is by far the largest. As a result, people are gradually shifting their vision to fisheye cameras, which can efficiently acquire 360° image information without blind spots [1][2][3][4] . This important characteristic gives it a wide range of application prospects, and it is widely used in intelligent surveillance, automatic driving, robotics, drones and other fields.…”