A Unified Detection Pipeline for Robust Object Detection in Fisheye-Based Traffic Surveillance
This work addresses the problem of degraded object detection near boundaries in fisheye imagery for traffic surveillance applications, representing an incremental improvement.
The paper tackled robust object detection in fisheye-based traffic surveillance by developing a unified detection pipeline with pre- and post-processing and ensemble strategies, achieving an F1 score of 0.6366 and placing 8th out of 62 teams in the 2025 AI City Challenge Track 4.
Fisheye cameras offer an efficient solution for wide-area traffic surveillance by capturing large fields of view from a single vantage point. However, the strong radial distortion and nonuniform resolution inherent in fisheye imagery introduce substantial challenges for standard object detectors, particularly near image boundaries where object appearance is severely degraded. In this work, we present a detection framework designed to operate robustly under these conditions. Our approach employs a simple yet effective pre and post processing pipeline that enhances detection consistency across the image, especially in regions affected by severe distortion. We train several state-of-the-art detection models on the fisheye traffic imagery and combine their outputs through an ensemble strategy to improve overall detection accuracy. Our method achieves an F1 score of0.6366 on the 2025 AI City Challenge Track 4, placing 8thoverall out of 62 teams. These results demonstrate the effectiveness of our framework in addressing issues inherent to fisheye imagery.