Causal neural set filtering for online multi-target tracking

Researchers proposed Causal Neural Set Filtering (CNSF), a neural set filter that encodes only current measurements while carrying past evidence in a structured recursive track state, reducing mean GOSPA and T-GOSPA relative to Track-MT3 by 19.3% and 30.4% on a held-out three-regime simulated test set. CNSF combines exclusive Sinkhorn association, association-conditioned Kalman-shaped updates with moment matching, and recurrent Bernoulli lifecycle modeling with measurement-driven birth, achieving 55.9% fewer parameters and a 3.76x speedup in single-thread CPU inference. The work targets redundant computation in Transformer-based multi-target tracking, where MT3/Track-MT3-style trackers repeatedly re-encode measurement windows.

arXiv:2609.16054v1 Announce Type: new Abstract: Transformer-based multi-target tracking MTT jointly learns data association and state estimation, but MT3/Track-MT3-style trackers repeatedly re-encode measurement windows, incurring redundant computation. We propose Causal Neural Set Filtering CNSF \footnote{\href{https://github.com/daihuangyu/CNSF}{Code: https://github.com/daihuangyu/CNSF}}, a neural set filter that encodes only current measurements while carrying past evidence in a structured recursive track state. CNSF combines exclusive Sinkhorn association, association-conditioned Kalman-shaped updates with moment matching, and recurrent Bernoulli lifecycle modeling with measurement-driven birth. These mechanisms impose soft one-to-one constraints, propagate association-induced state uncertainty, and support existence estimation under missed detections and birth--death transitions. On a held-out three-regime simulated test set, CNSF reduces mean GOSPA and T-GOSPA relative to Track-MT3 by 19.3\% and 30.4\%, with 55.9\% fewer parameters and a $3.76\times$ speedup in single-thread CPU inference.