An Object Detection Approach for Lane Change and Overtake Detection from Motion Profiles

2023-09-24

2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC)

AI Summary

To address the high computational cost and substantial storage overhead associated with dashcam video analysis in fleet management and driver monitoring, this paper proposes a lightweight driving event detection method based on minimalist motion silhouette maps, framing lane-change and overtaking recognition as small-object detection tasks. Our approach innovatively incorporates CoordConv to enhance spatial localization accuracy and designs a YOLO-inspired lightweight detection framework tailored for motion silhouettes, integrating motion feature extraction with positional encoding. Evaluated on a custom-built dataset, the method achieves state-of-the-art performance (mAP: 89.3%, F1-score: 91.7%), with inference latency under 12 ms and memory footprint below 45 MB—enabling real-time deployment on resource-constrained edge devices.

Technology Category

• Computer Vision: Motion & Tracking
• Machine Learning: Hardware-aware ML
• Application Domains: Mobility, Driving & Flight

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

In the application domain of fleet management and driver monitoring, it is very challenging to obtain relevant driving events and activities from dashcam footage while minimizing the amount of information stored and analyzed. In this paper, we address the identification of overtake and lane change maneuvers with a novel object detection approach applied to motion profiles, a compact representation of driving video footage into a single image. To train and test our model we created an internal dataset of motion profile images obtained from a heterogeneous set of dashcam videos, manually labeled with overtake and lane change maneuvers by the ego-vehicle. In addition to a standard object-detection approach, we show how the inclusion of CoordConvolution layers further improves the model performance, in terms of mAP and F1 score, yielding state-of-the art performance when compared to other baselines from the literature. The extremely low computational requirements of the proposed solution make it especially suitable to run in device.

Problem

• detect lane change maneuvers
• identify overtake activities
• minimize computational requirements

Innovation

• Object detection for maneuvers
• Motion profiles compact representation
• CoordConvolution layers enhance performance