<ul data-eligibleForWebStory="true"><li>Tracking multiple particles in noisy and cluttered scenes is challenging due to trajectory hypothesis combinatorial explosion.</li><li>The transformer architecture improves robustness but falls short in scenarios with a reduced set of trajectory hypotheses.</li><li>A hybrid approach combining self-attention of transformers with Bayesian filtering's reliability and interpretability is introduced.</li><li>Trajectory-to-detection association is done by solving a label prediction problem using a transformer encoder.</li><li>This hybrid approach prunes the hypothesis set, enabling efficient multiple-particle tracking in a Bayesian filtering framework.</li><li>The approach shows improved tracking accuracy and robustness against spurious detections in high clutter scenarios.</li></ul>

Attention-Bayesian Hybrid Approach to Modular Multiple Particle Tracking

Discover more