|
|
research:vslam:directtracker [2022/09/21 14:48] Mariia Gladkova created |
research:vslam:directtracker [2022/10/04 11:58] (current) Mariia Gladkova Add video and poster |
| |
Direct methods have shown excellent performance in the applications of visual odometry and SLAM. In this work we propose to leverage their effectiveness for the task of 3D multi-object tracking. To this end, we propose //DirectTracker//, a framework that effectively combines direct image alignment for the short-term tracking and sliding-window photometric bundle adjustment for 3D object detection. Object proposals are estimated based on the sparse sliding-window pointcloud and further refined using an optimization-based cost function that carefully combines 3D and 2D cues to ensure consistency in image and world space. We propose to evaluate 3D tracking using the recently introduced higher-order tracking accuracy (HOTA) metric and the generalized intersection over union similarity measure to mitigate the limitations of the conventional use of intersection over union for the evaluation of vision-based trackers. We perform evaluation on the KITTI Tracking benchmark for the Car class and show competitive performance in tracking objects both in 2D and 3D. | Direct methods have shown excellent performance in the applications of visual odometry and SLAM. In this work we propose to leverage their effectiveness for the task of 3D multi-object tracking. To this end, we propose //DirectTracker//, a framework that effectively combines direct image alignment for the short-term tracking and sliding-window photometric bundle adjustment for 3D object detection. Object proposals are estimated based on the sparse sliding-window pointcloud and further refined using an optimization-based cost function that carefully combines 3D and 2D cues to ensure consistency in image and world space. We propose to evaluate 3D tracking using the recently introduced higher-order tracking accuracy (HOTA) metric and the generalized intersection over union similarity measure to mitigate the limitations of the conventional use of intersection over union for the evaluation of vision-based trackers. We perform evaluation on the KITTI Tracking benchmark for the Car class and show competitive performance in tracking objects both in 2D and 3D. |
| |
| ===== Video ===== |
| |
| <html><center><iframe width="640" height="360" src="https://www.youtube.com/embed/hpm7bkGLdjs" frameborder="0" allowfullscreen></iframe></center></html> |
| |
| ===== Poster ===== |
| Our work is presented in the poster session of the 2nd General Assembly in [[https://www.mdsi.tum.de|the Munich Data Science Institute]] |
| {{ :research:vslam:directtracker:mdsi2022_directtracker_poster.jpg?direct&400 |}} |
| ===== Citation ===== |
| If you find our work useful, please consider citing: |
| <code> |
| @article{gladkova2022directtracker, |
| title={DirectTracker: 3D Multi-Object Tracking Using Direct Image Alignment and Photometric Bundle Adjustment}, |
| author={Gladkova, Mariia and Korobov, Nikita and Demmel, Nikolaus and O{\v{s}}ep, Aljo{\v{s}}a and Leal-Taix{\'e}, Laura and Cremers, Daniel}, |
| journal={arXiv preprint arXiv:2209.14965}, |
| year={2022} |
| } |
| </code> |
| |