3rd Workshop on Vehicle Retrieval in Surveillance (VRS 2015)
In video surveillance, the most important objects are people and vehicles. This workshop will focus on vehicles. Specifically, we will explore solutions to the practical problem of finding a given vehicle from an array of surveillance cameras using criteria such as the size, model, or color. Other potentially differentiating characteristics include whether a vehicle is a taxi, a delivery truck or a police car.
Furthermore, a search might be based on the location of the vehicle (i.e. north side of the highway, in the parking area behind the building) or the behavior of the vehicle (i.e. speed, direction of travel, whether the vehicle is parked illegally, or involvement in an accident or drive-by shooting.)
In the last few years, a number of studies have been undertaken to classify vehicles according to their type, make, model or color. However, the evaluation of each of these classification methods has been performed on in-house datasets. Since each dataset involves its own camera viewpoints, lighting conditions and resolutions, there has been no way to compare the relative merits of these methods.
Scope of the Workshop
In this workshop, we propose gathering together researchers who are working in this field, designing and building a public dataset, creating standardized evaluation criteria, and exploring new ideas in vehicle retrieval.
Areas of interest include but are not limited to:
- Vehicle color classification
- Vehicle type classification including type, make, model, logo, shape or size
- Vehicle attributes classification: type of grille, type of head/tail lights, presence of dome light etc.
- Vehicle behavior:
- Stopped at a given location for a specified time
- Direction of travel, speeding, making a u-turn, driving erratically etc.
- Anomalous behavior – not following the typical behavior for a given scene
- parked illegally, involved in an accident or drive-by shooting
- Matching vehicles across cameras, or based on queries into a gallery
- Vehicle detection and tracking
- Dr. Lisa M. Brown, IBM T.J. Watson Research Center
( e-mail: firstname.lastname@example.org )
- Prof. JunWei Hsieh, Taiwan National Ocean University
( e-mail: email@example.com )
Submission Deadlines (Extended!)
- Deadline for submission of electronic paper: June 15, 2015
- Notication of acceptance: June 22, 2015
- Deadline for camera-ready version: June 29, 2015
- Workshop in Karlsruhe: August 25, 2015
Paper Submission Guidelines
Papers should adhere to the AVSS2015 IEEE format using either:
LaTeX template here.
Word template here.
Detailed style and formatting instructions can be found in the templates. The blue numbering of text lines (the so-called ruler) must not be removed in the version for review, as it greatly facilitates the review process, but must be removed in the final version. Papers must be written in English and submitted in PDF format, each submission must be between 4 and 6 A4 pages long. o be anonymous, please read the instructions included in the paper templates.
Please submit your paper using the EasyChair submission.
A paper submission implies that, if the paper is accepted, one of the authors, or a proxy, will present the paper at the workshop. Submissions not using the above templates will be rejected without review. All accepted papers will be published in IEEE Xplore.