Neural and Neuromimetic Perception: A Comparative Study of Gender Classification from Human Gait
Neural and Neuromimetic Perception: A Comparative Study of Gender Classification from Human Gait
复制标题
DOI：
10.2352/j.percept.imaging.2020.3.1.010402
复制
发表时间：
2020-01
期刊：
影响因子：
--
通讯作者：
V. Sarangi;A. Pelah;W. Hahn;Elan Barenholtz
中科院分区：
文献类型：
--
作者：
V. Sarangi;A. Pelah;W. Hahn;Elan Barenholtz
摘要人类善于感知生物运动的目的，如性别歧视。观察员分类的性别步行者在显着以上的机会水平从点光分布的关节轨迹。然而，表现下降到机会水平或低于垂直反转的刺激，一种现象被称为反转效应。这种鲁棒性的缺乏可能反映了已经暴露于不充分的倒置刺激实例的通用学习机制，或者预先调整到直立刺激的专门机制的激活。为了解决这个问题，作者通过使用相同的生物运动刺激集，将人类的心理物理性能与神经模拟机器学习模型在步态性别分类中的计算性能进行了比较。实验结果表明，显着的相似性，其中包括在运动学运动线索的优势，结构线索的分类精度。其次，与人类一样，模型中的学习是在存在倒置效应的情况下表达的，这表明人类可能会使用通用学习系统来感知这项任务中的生物运动。最后，根据人类感知对模型进行修改，减轻了反演效果，提高了性能精度。该研究提出了一种从步态中研究人类性别感知的范式，并利用感知特征开发了一种鲁棒的人工步态分类器，用于临床运动分析等潜在应用。
Abstract Humans are adept at perceiving biological motion for purposes such as the discrimination of gender. Observers classify the gender of a walker at significantly above chance levels from a point-light distribution of joint trajectories. However, performance drops to chance level or below for vertically inverted stimuli, a phenomenon known as the inversion effect. This lack of robustness may reflect either a generic learning mechanism that has been exposed to insufficient instances of inverted stimuli or the activation of specialized mechanisms that are pre-tuned to upright stimuli. To address this issue, the authors compare the psychophysical performance of humans with the computational performance of neuromimetic machine-learning models in the classification of gender from gait by using the same biological motion stimulus set. Experimental results demonstrate significant similarities, which include those in the predominance of kinematic motion cues over structural cues in classification accuracy. Second, learning is expressed in the presence of the inversion effect in the models as in humans, suggesting that humans may use generic learning systems in the perception of biological motion in this task. Finally, modifications are applied to the model based on human perception, which mitigates the inversion effect and improves performance accuracy. The study proposes a paradigm for the investigation of human gender perception from gait and makes use of perceptual characteristics to develop a robust artificial gait classifier for potential applications such as clinical movement analysis.