| Home | Impressum | Sitemap | KIT


Environment Perception

Group Leader: Prof. Dr.-Ing. Stiller und Dr. rer. nat. Martin Lauer

Future cars will be able to drive autonomously without the help of a driver. This will reduce accidents and increase the comfort of driving. Thereto, the vehicles must be able to perceive and interpret their environment and to draw conclusions about the behavior of other traffic participants.

Within this research focus we are developing techniques for environment perception mainly based on monoscopic and stereoscopic cameras. The approaches deal with low level signal processing like efficient stereo matching and optical flow calculation, mid-level tasks like scene segmentation and reconstruction of scene geometry up to high-level scene interpretation tasks like behavior prediction for other traffic participants.

Turn assistance for truck Drivers

Accidents between trucks and bicycles are classified as extremely dangerous, often resulting in fatalities or heavy injuries. Enhancing the perception of the truck driver, this project aims to develop a camera and laser based bicyclist warning system in cooperation with the companies Spies and HFC, reducing the accident risk.

Contact: Johannes GräterWei Tian

Project Details

Pedestrian intention recognition for active safety Systems

In order to reduce the number of traffic accidents with pedestrians, we focus on active safety systems in cars. A vital component of those safety systems is the detection and prediction of motion of pedestrians. We utilize high resolution imaging for pedestrian detection and feature extraction. From this, we predict the future trajectories of pedestrians.

Contact: Eike Rehder

Project Details

Environment Perception to assist the Visually Impaired

In this field of research we develop assistance systems to help visually impaired people navigating public space. We are investigating systems based on stereoscopic cameras, that are carried unobtrusively on the blind person's head. This allows to extend the range of sensing from 1-2m – as possible with a white cane – to more than 10m.

Contact: Tobias Schwarze

Project Details

Object Detection for Inner City Traffic Scenarios

  Object detection and tracking for inner city traffic scenarios are challenging tasks for computer vision. This project deals with 3D object detection and optimal tracking strategies.

Contact: Philip Lenz

Project Details

Video Localization

We developed a method for high accuracy vision-only localization in a map of visual landmarks.

Contact: Henning Lategahn

Project Details

Camera Calibration

In this field of research we develop calibration routines for multi-camera setups with non-overlapping fields of view as well as special camera models for wide-angle and fisheye optics.

Contact: Tobias Strauß

Project Details

Environment Perception using Catadioptric Cameras

We use omnidirectional cameras on our vehicle to improve the field of view and capture a panoramic view of the environment. The project concentrates on calibrating stereo catadioptric systems for autonomous applications like egomotion estimation or 3D reconstruction.

Contact: Miriam Schönbein

Modelling of traffic situations at urban intersections

We are researching on a practical framework for modeling traffic situations at urban intersections, which can handle two problems: maneuver recognition and trajectory prediction of moving vehicles

Contact: Hong Quan Tran

Project Details

Parallel Processing for Real-time Stereo Vision

Stereo cameras are favorable sensors for 3D environmental perception, but processing their images in real-time is challenging. We therefore combine adaptive parallel computing on multi-core CPUs and GPUs with algorithms tailored for traffic scenes.

Contact: Benjamin Ranft 

Project Details     

Lane marking based localization and mapping

For vehicle localization we generate in a first (offline) step a map containing all visual lane markings. Onboard the vehicle we then match lane markings detected in the current camera images with the map to obain the position of the vehicle.

Contact: Markus Schreiber

Project Details



Environment Perception and Representation in Urban Environments

We work on environment representations, that can be used in modern driver assistance systems (DAS) as well as in assistance systems for visually impaired pedestrians.

Contact: Hannes Harms

Project Details

Lane track detection in urban Environments

Lane estimation of the ego vehicle plays a key role in navigating a car through unknown areas. In fact, solving this problem is a prerequisite for any vehicle driving autonomously in previously unmapped areas. In this project, the goal is to identify features in urban environments using low cost sensors and develop methods for estimating lanes based on these features.

Contact: Johannes Beck

Project Details


Other Activities