CAD/Graphics 2015Traffic situation visualization based on video composition
Graphical abstract
This is the place where to insert text.
Introduction
Vehicle detectors are usually distributed in-road to acquire spatiotemporal traffic statistic. The detectors record the number and the average speed of vehicles that pass within a time span. These data are important for city planners and traffic controllers [1] because the data depict overview and details of traffic situations over time. However, interpreting traffic statistic demands expertise and is not suitable for general populations. For example, the same driving speed in a countryside and an urban city could have very different experience because of light and heavy traffic densities. Providing users with a number of vehicles that pass is also unintuitive because it depends on the number of lanes on the road. Moreover, a light traffic flow may indicate few vehicles on the road or a serious traffic jam, which easily induces misleading. Accordingly, providing an interface for general users to realize traffic situations is essential.
Since transmitting videos captured from road surveillance cameras consumes expensive load, simulation techniques are presented for traffic visualization. The methods estimate velocity and density fields over the road network, followed by applying an agent-based traffic simulator to create 3D animations. They enjoy the visualization from various viewpoints and even allow users to observe traffic from a driver׳s perspective. However, a simulator cannot always realize the real traffic flows because driving behaviors are often different in countries and regions (Fig. 1), not to mention other conditions such as weather, rush hours, and holidays. To overcome this problem, we present an example-based system that visualizes traffic situations by synthesizing road surveillance videos.
Our goal is to visualize statistic VD data using real world materials. Specifically, we find a road segment where VD and surveillance camera are both available and extract the relations between them. For the place with a VD but without cameras, we synthesize a streaming video by composing the video clips in our database to visualize its traffic situations over time. The main advantage of this framework is generality. While a simulation technique is insufficient to create an animation that satisfies all driving behaviors, our system does not have this problem because they are already provided by road surveillance videos. Note that this framework also consumes light data transmission load because example videos are collected in advance. Only statistic VD data are transmitted when the traffic flows of a road segment are visualized.
The problems of traffic visualization in our framework are video retrieval and seamless composition. Considering the traffic situation of a video is unclear, to obtain the information, we seed particles on the video and track their motions. The number and speed of these particles that go outside the video coordinate or gather at the vanish point are recorded. After that, we compute a regression model to map the VD data and the particle flows so as to retrieve proper videos for composition when a macroscopic traffic statistic of another place is given. To prevent artifacts of video transition, we overlap consecutive videos by a number of frames and compute a surface that passes through pixels with the least distortion. Specifically, pixels on this surface should have small color variations and zero motions to avoid discontinuity artifacts and suddenly appearing or disappearing vehicles. We also apply the Poisson blending to smooth the difference of illumination conditions in videos for achieving high visual quality.
Our method synthesizes a streaming video to visualize the traffic situations of a place over time. This example-based framework can realize the traffic flows consistent with the sparse VD data at different regions. The main advantage of this framework is generality, which is able to visualize different traffic situations derived from driving behaviors, weathers, etc. We show the experimental results in Fig. 6, Fig. 7, and in the accompanying video to demonstrate the feasibility of our technique.
Section snippets
Related work
Traffic visualization: Visualizing traffic information is essential to city planners and traffic controllers. Many on-line services such as SigAlert and Google Maps depict the conditions of a traffic network by colorization. The abstracted visual means look clean and neat but lack details for understanding dynamic vehicle flows. Therefore, Walton et al. [2] projected live traffic videos onto maps to provide such information. However, the method consumes heavy transmission load when too many
Algorithm
Our goal is to visualize traffic statistic using videos captured by road surveillance cameras. To achieve the aim, the first step is to build the relations between traffic statistic and videos. We search for places where VD and surveillance camera are both available, and then compute a regression model to map the detected number of vehicles and average speed to the particle flows in a video. Specifically, our system cuts a streaming road surveillance video into short clips, with each clip
Results and discussions
We have implemented the presented approach using C and run the program on a desktop PC with Core i7 3.0 GHz CPU and GeForce GTX 550 GPU. The GPU based optical flow [25], [26] method is used to extract vehicle flows in a road surveillance video. We also applied the graph cut library [24] to compute the surface that can seamlessly transit one video to another. Finally, we apply the conjugate gradient method to solve the Poisson blending problem. This process is efficient because the hard transited
Conclusions
We have presented a system to visualize traffic situations by composing videos in a database. Given by macroscopic VD data that depicts the vehicles passing a road segment, our system retrieves proper videos that match the data for composition. It then reduces visual artifacts by computing a surface for seamless video transition and by solving a Poisson equation for illumination smoothing. Although the visualized traffic flows are of a fixed viewpoint, by collecting road surveillance videos
Acknowledgements
We thank the anonymous reviewers for their constructive comments. We are also grateful to Taiwan Area National Freeway Bureau for kindly providing us the dataset, and flickr users (Travel Aficionado and Sam Saunders) for sharing pictures with us. This work was supported in part by the Ministry of Science and Technology, Taiwan (Grant nos. 102-2221-E-009-083-MY3 and 101-2628-E-009-020-MY3).
References (27)
- et al.
Detailed traffic animation for urban road networks
Gr. Models
(2012) - et al.
Video-based personalized traffic learning
Gr Models
(2013) - Wilkie D, van den Berg JP, Lin MC, Manocha D. Self-aware traffic route planning. In: Association for the Advancement of...
- Walton S, Chen M, Ebert D. Livelayer—live traffic projection onto maps. In: Eurographics 2011—Posters, 2011. p....
- et al.
Virtualized trafficreconstructing traffic flows from discrete spatio-temporal data
2014 IEEE Virtual Real
(2009) - et al.
Continuum traffic simulation
(2010) - et al.
Virtualized trafficreconstructing traffic flows from discrete spatiotemporal data
IEEE Trans Vis Comput Gr
(2011) - et al.
Interactive hybrid simulation of large-scale traffic
ACM Trans Gr
(2011) - et al.
Flow reconstruction for data-driven traffic animation
ACM Trans Gr
(2013) - et al.
Vehicle pedestrian interaction for mixed traffic simulation
Comput Anim Virtual Worlds
(2015)
Graphcut texturesimage and video synthesis using graph cuts
ACM Trans Gr
Cited by (11)
Data-driven online traffic reconstructions: Interactively optimizing in virtual reality
2022, Computers and Graphics (Pergamon)Citation Excerpt :Wilkie et al. [21] reconstruct traffic flows from in-road sensor measurements generated data, which can be used to reconstruct traffic in real-time on large digital cityscape. Hsieh et al. [22] present an example-based system to visualize traffic situations by synthesizing road surveillance videos. Chao et al. [1] present a video-based approach to learn the driver’s specific driving characteristics offline for vividly reproducing the traffic flow.
Detection of vehicle tracks by a three-axis magnetometer
2018, Sensors and Actuators, A: PhysicalCitation Excerpt :For this purpose, a variety of sensing techniques have been applied to test vehicle motion parameters. Among them, the most widely used technique is video surveillance [1–3] by a camera probe which is expensive and needs manual operation. Additionally, this method has limited capability when the visibility is poor.
Shadow traffic: A unified model for abnormal traffic behavior simulation
2018, Computers and Graphics (Pergamon)Citation Excerpt :With the increasing volumes of traffic data and software tools capable of modeling urban scenes, numerous efforts have been devoted to traffic simulations. Since Sewall et al. first introduce the concept of “virtual traffic” [25], numerous detailed models have been proposed for realistic and efficient traffic simulations, including traffic flow descriptions [7–10,13,14,26,27], traffic flow reconstructions [11,12], and mixed traffic animations [28,29]. These models focus on physically based traffic simulations; that is, they aim to give a realistic description about the evolvement of traffic flows after an anomaly occurs.
- 1
Tel.: +886 989853715; fax: +886 3 5721490.