Semi-Automatic Framework for Traffic Landmark Annotation

Abstract

We present a semi-automatic annotation method to build a large dataset for traffic landmark detection, where traffic landmarks include traffic signs, traffic lights as well as road markings. Labor-intensive bounding box tagging is a huge challenge to generate a large dataset for detection algorithms. To mitigate the labor, we adopt a high-definition (HD) map and a positioning system. We propose a process to align the HD map and images semi-automatically. Through the registration, the annotations of the HD map can be directly tagged onto traffic landmarks in the images. To make full use of the HD map for the dataset generation, we annotate the traffic landmarks with reference points, following the way that they are represented in the HD map, instead of the bounding boxes. The proposed semi-automatic method speeds up the annotation by a factor of 3.19, as compared to the manual annotation. Our dataset consists of about 150,000 images and includes about 470,000 annotated traffic landmarks. We train a deep neural network on our dataset to detect the traffic landmarks, and its performance is evaluated using a novel evaluation metric. Moreover, we show that the pretrained traffic landmark detection network is effective in detecting traffic landmarks in other countries using the bounding box by fine-tuning.

An innovative traffic light recognition method using vehicular ad-hoc networks

“…A training neural network for traffic light detection was used; however, it lacked tracking. Global Navigation Satellite System (GNSS) and inertial navigation system (INS) are integrated, which increases the cost of cars 14 . Numerous samples of traffic lights were randomly identified as positive samples, and many samples as negative samples were identified among the traffic light samples (non-traffic light samples).…”

“…Complex detection methods are used in state-of-the-art detection procedures. The camera was installed on the dashboard and windshield in the front direction in CNN, ICFT, and semi-automatic traffic light detection methods 14,15 . The video set consists of several frames, each of which contains many instances of various traffic signals (red, green, and yellow).…”

Traffic Landmark Matching Framework for HD-Map Update: Dataset Training Case Study

“…Precision is defined as the ratio of the total number of true predictions to the total number of predictions. The IoU threshold determines whether the prediction is true or false [8]. Precision and recall are defined in Equations ( 1) and ( 2), and Equations ( 3) and ( 4) define the accuracy and F1 score, as given below.…”

“…High detection rates and precision localization are essential for the automation of HD map updates. Commercial map providers, such as Mobileye [4], HERE [5], CAR-MERA [6], and TomTom [7], have already produced HD maps annotated with 3D geographic and semantic information on traffic landmarks with an accuracy of 10 cm [8].…”

“…Moreover, Lee et al [8] proposed a semi-automatic method that speeds up the annotation by a factor of 3.19 in comparison to manual annotation. The dataset consists of approximately 150,000 images and includes approximately 470,000 annotated traffic landmarks.…”