Occlusion-Aware Multi-Object Tracking via Expected Probability of Detection
Jan Krejčí, Oliver Kost, Yuxuan Xia, Lennart Svensson, Ondřej Straka · 2025
Read Occlusion-Aware Multi-Object Tracking via Expected Probability of Detection by Jan Krejčí, Oliver Kost, Yuxuan Xia, Lennart Svensson, Ondřej Straka (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-20239.txt.
Description
Krejčí, Kost, Xia, Svensson, and Straka teach trackers to expect objects to hide each other. Their 2025 paper handles occlusion in multi-object tracking. Sensors need line of sight to spot anything. Objects in front block those behind them. Trackers that ignore this lose hidden objects entirely.
The standard detection model assumes a fixed spotting chance. Their upgrade makes detection probability depend on all objects present. Each object gets an expected probability of detection. The expectation conditions on that object existing. It uses the reduced Palm density from point process theory. Visibility is judged relative to the sensor and the crowd. Uncertainties about every object enter one clear formula. A multi-Bernoulli mixture filter with marks implements it. Visual tracking tests showed the strategy beating rivals. Safety uses like driving and naval watch motivate the care. The method stays fully model-based throughout.
Catalog record
| CREATOR | Jan Krejčí, Oliver Kost, Yuxuan Xia, Lennart Svensson, Ondřej Straka |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 96 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-20239