PEOPLE

Research Teams

Research Teams

PEOPLE > Research Teams

Dongseok Kim

3-4-1 Project Co-director

  • Affiliation CISS
  • Tel +82-2-3498-7535
  • E-mail

    kds1130@kaist.ac.kr

  • Homepage

  • Research Topic Vision-based object detection
Project : Vision-based pedestrian detection for automotive application
Purpose ○ This project targets to develop vision-based pedestrian detection algorithms using vehicle-mounted rear-view camera (backup or black box) to prevent back-over accidents
Contents ○ Wide-angle lens camera-based pedestrian detection
- Development of regions of interest generation method specific for camera parameters, such as height of camera, field of view, heading angle
- Research on feature extraction method for wide-angle images by comparison of existing feature extraction methods

○ Pedestrian detection with visible and depth image fusion
- Development of regions of interest generation utilizing depth information
- Development of depth image-specific feature extraction method
- Research on multi-source/multi-feature based pedestrian classification method

○ Research on risk assessment for detecting pedestrians
Expected Contribution ○ Reducing back-over crashes of pedestrians
○ Secure core technologies for camera-based adaptive drive assistance systems
○ Expansion of person detection technology to other smart systems
Publication
No Title Year Phase
 
Patent
No Title Country Date Phase
 
Conference
No title Conference Name Date Phase
 
맨위로

Warning: Unknown: Your script possibly relies on a session side-effect which existed until PHP 4.2.3. Please be advised that the session extension does not consider global variables as a source of data, unless register_globals is enabled. You can disable this functionality and this warning by setting session.bug_compat_42 or session.bug_compat_warn to off, respectively in Unknown on line 0