Monocular Cyclist Detection with Convolutional Neural Networks

TL;DR

This work tackles the risk of truck–cyclist collisions in blind spots by developing real-time monocular cyclist detection using lightweight CNNs (EfficientDet Lite and SSD MobileNetV2). A newly created auxiliary dataset (~$2\times 10^{4}$ images, ~40k cyclist instances) enables high performance, with $mAP$ values around $0.956$ for EfficientDet Lite on $IoU=0.5$ and fast on‑device inferences on the Google Coral Dev Board around $15$ ms. The system is packaged into a portable blind‑spot device with a 7‑inch display and alerts, achieving real‑time capability at cab scale and costing roughly $200$, demonstrating strong feasibility for field deployment. The results suggest meaningful safety benefits for reducing right‑hook truck–cyclist collisions and lay groundwork for broader industry testing, integration with additional sensors, and further improvements using advanced architectures.

Abstract

Cycling is an increasingly popular method of transportation for sustainability and health benefits. However, cyclists face growing risks, especially when encountering large vehicles on the road. This study aims to reduce the number of vehicle-cyclist collisions, which are often caused by poor driver attention to blind spots. To achieve this, we designed a state-of-the-art real-time monocular cyclist detection that can detect cyclists with object detection convolutional neural networks, such as EfficientDet Lite and SSD MobileNetV2. First, our proposed cyclist detection models achieve greater than 0.900 mAP (IoU: 0.5), fine-tuned on a newly proposed cyclist image dataset comprising over 20,000 images. Next, the models were deployed onto a Google Coral Dev Board mini-computer with a camera module and analyzed for speed, reaching inference times as low as 15 milliseconds. Lastly, the end-to-end cyclist detection device was tested in real-time to model traffic scenarios and analyzed further for performance and feasibility. We concluded that this cyclist detection device can accurately and quickly detect cyclists and has the potential to improve cyclist safety significantly. Future studies could determine the feasibility of the proposed device in the vehicle industry and improvements to cyclist safety over time.

Figures (13)

• Figure 1: Semi-trailer Truck Right-Hook Turn Scenario.
• Figure 2: Limited Visibility of Cyclists From a Semi-trailer Truck Driver.
• Figure 3: Overview of Cyclist Detection Device for Semi-trailer Truck Blind Spots.
• Figure 4: Right-Hook Turning Scenario v.s. Model Testing Environment.
• Figure 5: Example Images From the Newly Proposed Cyclist Dataset.