返回
An Uncertainty-Aware Performance Measure for Multi-Object Tracking
DOI:10.1109/LSP.2021.3103488.png)
摘要
En 中文
Evaluating the performance of multi-object tracking (MOT) methods is not straightforward, and existing performance measures fail to consider all the available uncertainty information in the MOT context. This can lead practitioners to select models which produce uncertainty estimates of lower quality, negatively impacting any downstream systems that rely on them. Additionally, most MOT performance measures have hyperparameters, which makes comparisons of different trackers less straightforward. We propose the use of the negative log-likelihood (NLL) of the multi-object posterior given the set of ground-truth objects as a performance measure. This measure takes into account all available uncertainty information in a sound mathematical manner without hyperparameters. We provide efficient algorithms for approximating the computation of the NLL for several common MOT algorithms, show that in some cases it decomposes and approximates the widely-used GOSPA metric, and provide several illustrative examples highlighting the advantages of the NLL in comparison to other MOT performance measures.
Keyword:
Uncertainty
Measurement uncertainty
Measurement
Signal processing algorithms
Density measurement
Approximation algorithms
Task analysis
Multitarget tracking
multi-object tracking
performance measure
uncertainty evaluation
期刊
IF:
9.6
论文数:
1.1W
被引数:
1.7W
机构
引用论文
Message Passing Algorithms for Scalable Multitarget Tracking可扩展多目标跟踪的消息传递算法
PROCEEDINGS OF THE IEEE
IF25.9
Health Impact Modelling of Active Travel Visions for England and Wales Using an Integrated Transport and Health Impact Modelling Tool (ITHIM)
PLoS ONE
IF0
Antibodies to phosphatidylserine/prothrombin complex in suspected antiphospholipid syndrome in the absence of antibodies to cardiolipin or Beta-2-glycoprotein I
Lupus
IF0

