Visual Motion of Vehicle Interaction for Autonomous Driving

<p>Vehicles move on roads by performing regular actions on the ground according to road environments and traffic rules. Such motion of surrounding vehicles is captured by vehicle cameras on the ego-vehicle in the video. According to the visual appearance of vehicles and their motion trajectories, human drivers can identify the interactions of surrounding vehicles during stable vehicle action/driving without difficulty.</p>

<p>This dataset collects visual data of driving scenes in videos. We label the qualitative vehicle actions around the camera based on the detailed motion trajectories of surrounding vehicles. The data are used in deep learning of temporal events and vehicle behaviors in the sensing of autonomous vehicles.</p>

<p>Vehicle interactions are limited to the range of 50m. Depending on different lanes and relative speed around the ego-vehicle, 14 classes of qualitative motion (interaction) are specified. They are identified in the video for preparing ego-vehicle action and avoiding collision. This is particularly important for fast-moving vehicles in order to plan vehicle motion and path ahead and respond promptly in accident avoidance.</p>

<p>To obtain the motion of vehicles moving on a horizontal plane, the video is sampled at the height of the horizon to produce the Motion Profile at 30fps temporally, which contains the trajectories of vehicles and background scenes. Along with the Motion Profile, the width of the vehicle in the images, which reflects the vehicle depth, is also obtained and plotted with the Motion Profile, as the input of a spatial-temporal deep network to understand vehicle interactions.</p>

<p>The motion trajectories are annotated with the interaction classes in specified colors, separated at the time moment when a new interaction starts.</p>

<p>Please see the details in PDF in the dataset.</p>