English

Joint Spatial-Temporal and Appearance Modeling with Transformer for Multiple Object Tracking

Computer Vision and Pattern Recognition 2022-06-01 v1

Abstract

The recent trend in multiple object tracking (MOT) is heading towards leveraging deep learning to boost the tracking performance. In this paper, we propose a novel solution named TransSTAM, which leverages Transformer to effectively model both the appearance features of each object and the spatial-temporal relationships among objects. TransSTAM consists of two major parts: (1) The encoder utilizes the powerful self-attention mechanism of Transformer to learn discriminative features for each tracklet; (2) The decoder adopts the standard cross-attention mechanism to model the affinities between the tracklets and the detections by taking both spatial-temporal and appearance features into account. TransSTAM has two major advantages: (1) It is solely based on the encoder-decoder architecture and enjoys a compact network design, hence being computationally efficient; (2) It can effectively learn spatial-temporal and appearance features within one model, hence achieving better tracking accuracy. The proposed method is evaluated on multiple public benchmarks including MOT16, MOT17, and MOT20, and it achieves a clear performance improvement in both IDF1 and HOTA with respect to previous state-of-the-art approaches on all the benchmarks. Our code is available at \url{https://github.com/icicle4/TranSTAM}.

Keywords

Cite

@article{arxiv.2205.15495,
  title  = {Joint Spatial-Temporal and Appearance Modeling with Transformer for Multiple Object Tracking},
  author = {Peng Dai and Yiqiang Feng and Renliang Weng and Changshui Zhang},
  journal= {arXiv preprint arXiv:2205.15495},
  year   = {2022}
}
R2 v1 2026-06-24T11:33:55.840Z