Pedestrian detection is the cornerstone of many vision based applications, starting from object tracking to video surveillance and more recently, autonomous driving. With the rapid development of deep learning in object detection, pedestrian detection has achieved very good performance in traditional single-dataset training and evaluation setting. However, in this study on generalizable pedestrian detectors, we show that, current pedestrian detectors poorly handle even small domain shifts in cross-dataset evaluation. We attribute the limited generalization to two main factors, the method and the current sources of data. Regarding the method, we illustrate that biasness present in the design choices (e.g anchor settings) of current pedestrian detectors are the main contributing factor to the limited generalization. Most modern pedestrian detectors are tailored towards target dataset, where they do achieve high performance in traditional single training and testing pipeline, but suffer a degrade in performance when evaluated through cross-dataset evaluation. Consequently, a general object detector performs better in cross-dataset evaluation compared with state of the art pedestrian detectors, due to its generic design. As for the data, we show that the autonomous driving benchmarks are monotonous in nature, that is, they are not diverse in scenarios and dense in pedestrians. Therefore, benchmarks curated by crawling the web (which contain diverse and dense scenarios), are an efficient source of pre-training for providing a more robust representation. Accordingly, we propose a progressive fine-tuning strategy which improves generalization. Code and models cab accessed at https://github.com/hasanirtiza/Pedestron.
翻译:Pedestrian探测是许多基于视觉的应用程序的基石,从物体跟踪到视频监视,以及最近自主驾驶。随着在物体探测方面的深层学习的迅速发展,行人探测在传统的单一数据集培训和评估环境中取得了非常好的成绩。然而,在这项关于通用行人探测器的研究中,我们表明,目前行人探测器在交叉数据集评价方面处理得很差,即使是小域的移动也处理不力。我们把这种有限的概括性归因于两个主要因素,即方法和目前的数据来源。关于方法,我们说明目前行人探测器的设计选择(如锚定设置)中的偏差是有限一般化的主要促成因素。大多数现代行人探测器是专门为目标数据集设计的,在传统的单一培训和测试管道中确实取得很高的性能,但在通过交叉数据集评估进行评估时,工作业绩会下降。因此,一般物体探测器在交叉数据集评价方面表现得比艺术行人探测器的状态要好,因为其通用设计不同。关于数据,我们表明自主驾驶基准在性质上是单独独一的,因此,它们并不具有精确性,在传统的单一性,因此,在传统的单项培训和测试管道中,在高端探测器中并不具有高度的标志性定型,因此,在高的轨道上,在高的模型中,我们根据地标定标定的模型中提供了一种高的模型。我们根据地标定式地标定,在高的路径上,在高的路径上,在高的模型。我们,在高的路径上,在高的模型。我们,在高的路径上,在高的模型中,在高的路径上,在高。我们标定标定标定标,在高。我们,在高。我们,在高。我们根据地标,在高的轨道上,从上,在高的精确性地标上,在高的深度标,在高,从上,从上,从上,从上,从上,从上,在高。在高,在高路标定地标定地标定地标上,在高的精确标上,在高。