Institute of Robotics and Machine Intelligence, Poznan University of Technology, 60-965 Poznan, Poland.
Sensors (Basel). 2022 Jan 30;22(3):1082. doi: 10.3390/s22031082.
Detecting pedestrians in autonomous driving is a safety-critical task, and the decision to avoid a a person has to be made with minimal latency. Multispectral approaches that combine RGB and thermal images are researched extensively, as they make it possible to gain robustness under varying illumination and weather conditions. State-of-the-art solutions employing deep neural networks offer high accuracy of pedestrian detection. However, the literature is short of works that evaluate multispectral pedestrian detection with respect to its feasibility in obstacle avoidance scenarios, taking into account the motion of the vehicle. Therefore, we investigated the real-time neural network detector architecture You Only Look Once, the latest version (YOLOv4), and demonstrate that this detector can be adapted to multispectral pedestrian detection. It can achieve accuracy on par with the state-of-the-art while being highly computationally efficient, thereby supporting low-latency decision making. The results achieved on the KAIST dataset were evaluated from the perspective of automotive applications, where low latency and a low number of false negatives are critical parameters. The middle fusion approach to YOLOv4 in its Tiny variant achieved the best accuracy to computational efficiency trade-off among the evaluated architectures.
在自动驾驶中检测行人是一项安全关键任务，必须以最小的延迟做出避免行人的决策。广泛研究了结合 RGB 和热图像的多光谱方法，因为它们可以在不同的光照和天气条件下实现鲁棒性。采用深度神经网络的最新解决方案提供了行人检测的高精度。然而，文献中缺乏关于多光谱行人检测在考虑车辆运动的避障场景中的可行性的评估工作。因此，我们研究了实时神经网络检测器架构 You Only Look Once，最新版本 (YOLOv4)，并证明可以将其应用于多光谱行人检测。它可以在具有高度计算效率的同时实现与最先进水平相当的准确性，从而支持低延迟决策。从汽车应用的角度评估了在 KAIST 数据集上获得的结果，其中低延迟和低误报率是关键参数。在其 Tiny 变体中，YOLOv4 的中间融合方法在评估的架构中实现了最佳的准确性与计算效率的权衡。
