FusionSORT: Fusion Methods for Online Multi-object Visual Tracking
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910935367548928 |
|---|---|
| author | Baisa, Nathanael L. |
| author_facet | Baisa, Nathanael L. |
| contents | In this work, we investigate four different fusion methods for associating detections to tracklets in multi-object visual tracking. In addition to considering strong cues such as motion and appearance information, we also consider weak cues such as height intersection-over-union (height-IoU) and tracklet confidence information in the data association using different fusion methods. These fusion methods include minimum, weighted sum based on IoU, Kalman filter (KF) gating, and hadamard product of costs due to the different cues. We conduct extensive evaluations on validation sets of MOT17, MOT20 and DanceTrack datasets, and find out that the choice of a fusion method is key for data association in multi-object visual tracking. We hope that this investigative work helps the computer vision research community to use the right fusion method for data association in multi-object visual tracking. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_00843 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FusionSORT: Fusion Methods for Online Multi-object Visual Tracking Baisa, Nathanael L. Computer Vision and Pattern Recognition In this work, we investigate four different fusion methods for associating detections to tracklets in multi-object visual tracking. In addition to considering strong cues such as motion and appearance information, we also consider weak cues such as height intersection-over-union (height-IoU) and tracklet confidence information in the data association using different fusion methods. These fusion methods include minimum, weighted sum based on IoU, Kalman filter (KF) gating, and hadamard product of costs due to the different cues. We conduct extensive evaluations on validation sets of MOT17, MOT20 and DanceTrack datasets, and find out that the choice of a fusion method is key for data association in multi-object visual tracking. We hope that this investigative work helps the computer vision research community to use the right fusion method for data association in multi-object visual tracking. |
| title | FusionSORT: Fusion Methods for Online Multi-object Visual Tracking |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2501.00843 |