Human behaviour recognition in data-scarce domains
Human behaviour recognition in data-scarce domains
复制标题
DOI：
10.1016/j.patcog.2015.02.019
复制
发表时间：
2015-08
期刊：
影响因子：
--
通讯作者：
R. Baxter;N. Robertson;D. Lane
中科院分区：
文献类型：
--
作者：
R. Baxter;N. Robertson;D. Lane
本文提出了一种新的理论，用于执行多智能体活动识别，而不需要大量的训练语料库。对数据的需求减少意味着可以在传统上无法获得注释数据集的领域内执行鲁棒的概率识别。复杂的人类活动是由一系列潜在的原始活动组成的。我们不假设基元的精确时间顺序是必要的，因此可以使用无序的包来表示复杂的活动。我们的三层架构包括低级别的视频跟踪，事件分析和高级推理。高级推理使用Rao-Blackwellised粒子滤波器的新级联扩展来执行。模拟退火用于识别参与多代理活动的代理对。我们验证我们的框架使用基准PETS 2006视频监控数据集和我们自己的序列，并实现了平均识别F-得分为0.82。我们的方法比隐马尔可夫模型基线平均提高了17%。
This paper presents the novel theory for performing multi-agent activity recognition without requiring large training corpora. The reduced need for data means that robust probabilistic recognition can be performed within domains where annotated datasets are traditionally unavailable. Complex human activities are composed from sequences of underlying primitive activities. We do not assume that the exact temporal ordering of primitives is necessary, so can represent complex activity using an unordered bag. Our three-tier architecture comprises low-level video tracking, event analysis and high-level inference. High-level inference is performed using a new, cascading extension of the Rao–Blackwellised Particle Filter. Simulated annealing is used to identify pairs of agents involved in multi-agent activity. We validate our framework using the benchmarked PETS 2006 video surveillance dataset and our own sequences, and achieve a mean recognition F-Score of 0.82. Our approach achieves a mean improvement of 17% over a Hidden Markov Model baseline.