FAMNet: Joint Learning of Feature, Affinity and Multi-dimensional Assignment for Online Multiple Object Tracking
About
Data association-based multiple object tracking (MOT) involves multiple separated modules processed or optimized differently, which results in complex method design and requires non-trivial tuning of parameters. In this paper, we present an end-to-end model, named FAMNet, where Feature extraction, Affinity estimation and Multi-dimensional assignment are refined in a single network. All layers in FAMNet are designed differentiable thus can be optimized jointly to learn the discriminative features and higher-order affinity model for robust MOT, which is supervised by the loss directly from the assignment ground truth. We also integrate single object tracking technique and a dedicated target management scheme into the FAMNet-based tracking system to further recover false negatives and inhibit noisy target candidates generated by the external detector. The proposed method is evaluated on a diverse set of benchmarks including MOT2015, MOT2017, KITTI-Car and UA-DETRAC, and achieves promising performance on all of them in comparison with state-of-the-arts.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Multiple Object Tracking | MOT17 (test) | MOTA52 | 921 | |
| Multi-Object Tracking | MOT17 | MOTA52 | 55 | |
| Multi-Object Tracking | KITTI (test) | -- | 51 | |
| Multi-Object Tracking | MOT17 1.0 (test) | MOTA52 | 48 | |
| Multiple Object Tracking | MOT 17 (public detections) | MOTA52 | 15 | |
| Multi-Object Tracking | KITTI Tracking Car | HOTA52.6 | 10 | |
| Multiple Object Tracking | UA-DETRAC v1 (test) | MOTA19.8 | 8 | |
| Multiple Object Tracking | UA-DETRAC | MOTA19.8 | 8 |