Extracting Driving Behavior: Global Metric Localization from Dashcam Videos in the Wild
Given the advance of portable cameras, many vehicles are equipped with always-on cameras on their dashboards (referred to as dashcam). We aim to utilize these dashcam videos harvested in the wild to extract the driving behavior—global metric localization
- PDF / 5,153,450 Bytes
- 13 Pages / 439.37 x 666.142 pts Page_size
- 51 Downloads / 236 Views
2
Departmant of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan [email protected], [email protected], [email protected] Departmant of Computer Science, National Tsing Hua University, Hsinchu, Taiwan [email protected], [email protected], [email protected]
Abstract. Given the advance of portable cameras, many vehicles are equipped with always-on cameras on their dashboards (referred to as dashcam). We aim to utilize these dashcam videos harvested in the wild to extract the driving behavior—global metric localization of 3D vehicle trajectories (Fig. 1). We propose a robust approach to (1) extract a relative vehicle 3D trajectory from a dashcam video, (2) create a global metric 3D map using geo-localized Google StreetView RGBD panoramic images, and (3) align the relative vehicle 3D trajectory to the 3D map to achieve global metric localization. We conduct an experiment on 50 dashcam videos captured in 11 cities under various traffic conditions. For each video, we uniformly sample at least 15 control frames per road segment to manually annotate the ground truth 3D locations of the vehicle. On control frames, the extracted 3D locations are compared with these manually labeled ground truths to calculate the distance in meters. Our proposed method achieves an average error of 2.05 m and 85.5 % of them have error no more than 5 m. Our method significantly outperforms other vision-based baseline methods and is a more accurate alternative method than the most widely used consumer-level Global Positioning System (GPS).
Keywords: Camera localization
1
· Structure from motion
Introduction
Recently, self-driving car is one of the hottest topic in computer vision and it has received a huge amount of industrial investment to solve this holy-grail problem. One very important topic for advancing self-driving car is to build a realistic simulation environment, in particular, realistic driving behavior of other AI agents in the environment. Collecting in-house driving behavior data from humans is time-consuming and not scalable to cover many corner cases. Hence, c Springer International Publishing Switzerland 2016 G. Hua and H. J´ egou (Eds.): ECCV 2016 Workshops, Part I, LNCS 9913, pp. 136–148, 2016. DOI: 10.1007/978-3-319-46604-0 10
Global Metric Localization from Dashcam Videos in the Wild Input
137
Output Bird’s eye view
Image
driving direction Datong Road
StreetView Depth
Global Metric Loco.
Dashcam frame
Rendered view
Dashcam
Fig. 1. Global Metric Localization of 3D vehicle trajectory. Left-Panel: inputs including StreetView RGBD panorama images (Top) and dashcam frames (Bottom). RightPanel: output—3D trajectory (yellow dots) in bird’s eye view (Top). A rendered view compared to a dashcam frame is shown at the Bottom. (Color figure online)
we propose to crowd-source driving behavior from many individual drivers using cheap portable cameras. Thanks to the advance of portable camera, many vehicles are equipped with always-on cameras on their da
Data Loading...
