Stixel-based Traffic Scene Representation Using U-disparity

The Stixel World is a valid approach of interpreting 3D traffic scenes. This paper presents a novel and simple method to establish the Stixel World by exploiting the properties of U-disparity. Unlike the conventional stixel extraction method, the proposed method does not model the ground plane, but construct a U-disparity map and accordingly remove ground-related points and background-related points in virtue of U-disparity properties. The resulting U-disparity map is re-projected back to the disparity map so that the base and top-points of foreground obstacles can be determined. Experiments show that the proposed method can effectively build a Stixel World for a variety of scenarios and is not constrained by topography and object classes.

[1]  Markus Enzweiler,et al.  Efficient Stixel-based object recognition , 2012, 2012 IEEE Intelligent Vehicles Symposium.

[2]  Uwe Franke,et al.  From stixels to objects — A conditional random field based approach , 2013, 2013 IEEE Intelligent Vehicles Symposium (IV).

[3]  Jean-Philippe Tarel,et al.  Real time obstacle detection in stereovision on non flat road geometry through "v-disparity" representation , 2002, Intelligent Vehicle Symposium, 2002. IEEE.

[4]  Reinhard Klette,et al.  Improved Stixel Estimation Based on Transitivity Analysis in Disparity Space , 2017, CAIP.

[5]  Andreas Geiger,et al.  Are we ready for autonomous driving? The KITTI vision benchmark suite , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  Lounis Chermak,et al.  Stixel based scene understanding for autonomous vehicles , 2017, 2017 IEEE 14th International Conference on Networking, Sensing and Control (ICNSC).

[7]  Luc Van Gool,et al.  Stixels estimation without depth map computation , 2011, 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops).

[8]  Uwe Franke,et al.  Efficient representation of traffic scenes by means of dynamic stixels , 2010, 2010 IEEE Intelligent Vehicles Symposium.

[9]  Uwe Franke,et al.  Low-level fusion of color, texture and depth for robust road scene understanding , 2015, 2015 IEEE Intelligent Vehicles Symposium (IV).

[10]  Uwe Franke,et al.  The Stixel World - A Compact Medium Level Representation of the 3D-World , 2009, DAGM-Symposium.

[11]  Ethan Fetaya,et al.  StixelNet: A Deep Convolutional Network for Obstacle Detection and Road Segmentation , 2015, BMVC.

[12]  Keiichi Uchimura,et al.  A complete U-V-disparity study for stereovision based 3D driving environment analysis , 2005, Fifth International Conference on 3-D Digital Imaging and Modeling (3DIM'05).

[13]  Peter H. N. de With,et al.  Extending the Stixel World with online self-supervised color modeling for road-versus-obstacle segmentation , 2014, 17th International IEEE Conference on Intelligent Transportation Systems (ITSC).