Diverse Large-Scale ITS Dataset Created from Continuous Learning for Real-Time Vehicle Detection

TL;DR

This paper introduces a large-scale, diverse vehicle detection dataset created through continuous learning and crowd-sourced data mining, enabling real-time detection with high accuracy in complex traffic environments.

Contribution

It presents a novel cloud-based positive and negative mining process combined with a large-scale learning system for traffic event detection, significantly improving dataset diversity and detection accuracy.

Findings

Achieved at least 95% accuracy on test data

Trained on 1 million Haar-like features from 70,000 frames

System scales to over one billion annotated frames by 2015

Abstract

In traffic engineering, vehicle detectors are trained on limited datasets resulting in poor accuracy when deployed in real world applications. Annotating large-scale high quality datasets is challenging. Typically, these datasets have limited diversity; they do not reflect the real-world operating environment. There is a need for a large-scale, cloud based positive and negative mining (PNM) process and a large-scale learning and evaluation system for the application of traffic event detection. The proposed positive and negative mining process addresses the quality of crowd sourced ground truth data through machine learning review and human feedback mechanisms. The proposed learning and evaluation system uses a distributed cloud computing framework to handle data-scaling issues associated with large numbers of samples and a high-dimensional feature space. The system is trained using AdaBoost on $1,000,000$ Haar-like features extracted from $70,000$ annotated video frames. The trained real-time vehicle detector achieves an accuracy of at least $95\%$ for $1/2$ and about $78\%$ for $19/20$ of the time when tested on approximately $7,500,000$ video frames. At the end of 2015, the dataset is expect to have over one billion annotated video frames.