Demos on road scene understanding and autonomous driving

 

Object tracking from moving platform (new) for Velodyne HDL64 sensor

Preliminary demo, to be published soon, 2016

Object recognition with deep learning (new) for Velodyne HDL64 sensor

Preliminary demo, to be published soon, 2016

Crossmodal Point Cloud Registration for Mobile Laser Scanning Data (new)

tested with Velodyne HDL64, VLP16 and Riegl VMX450 sensors

Published @ International Conference on Pattern Recognition 2016

GPS/IMU free SLAM for Velodyne HDL64 and VLP16 sensors (new)

Published @ International Conference on Pattern Recognition 2016

A Model-based Approach for Fast Vehicle Detection in Continuously Streamed Urban LIDAR Point Clouds

Published @ Scene Understanding for Autonomous Systems at ACCV'14

Fast 3-D Urban Object Detection on Streaming Point Cloud: we present a simple, yet efficient hierarchical grid data structure and corresponding algorithms that significantly improve the processing speed of the object detection task.

Published @ Road Scene Understanding and Autonomous Driving at ECCV'14

Real time point cloud segmentation for Velodyne HDL64 and VLP16 (new!!!) sensors: are able to automatically interpret the LIDAR point cloud stream obtained from a moving platform, segment different point cloud classes,

Published @ ISPRS Workshop on 3D Virtual City Modeling, 2013

Zebra crossing and street object detection we perform real-time localization and identification of typical urban objects, such as traffic signs, vehicles or crosswalks.

Published @ CogInfoCom 2013,

Point cloud filtering: vegetation detection, point cloud enhancement, and moving-static object separation

Published @ IEEE Int'l Workshop on Content-Based Multimedia Indexing, 2013

Data aquisition point cloud sequences are obtained from a moving vehicle, using a Velodyne HDL-64E Rotating Multi-Beam LIDAR sensor.