当前位置: X-MOL 学术IEEE Robot. Automation Lett. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A 500-fps Pan-tilt Tracking System with Deep-learning-based Object Detection
IEEE Robotics and Automation Letters ( IF 4.6 ) Pub Date : 2021-04-01 , DOI: 10.1109/lra.2020.3048653
Mingjun Jiang , Kohei Shimasaki , Shaopeng Hu , Taku Senoo , Idaku Ishii

In this letter, we propose a fast mirror-drive pan-tilt target tracking system that can robustly track an object whose appearance varies in a complex background at 500 fps. By assuming a small image displacement between frames, which is a property of high-frame rate vision, we develop an fast object tracking algorithm by hybridizing the convolutional-neural-network (CNN) based object detection with template-matching (TM) based tracking operating at hundreds of frames per second (fps). For object tracking with high-speed visual feedback, the proposed tracking algorithm can remarkably reduce dozens-of-milliseconds-latency in the CNN-based object detection by simultaneously executing TM-based tracking for several images at consecutive frames within a few milliseconds. In the proposed pan-tilt tracking system, when the current tracked objects are occluded or out of the camera view, it can recognize objects to be newly tracked with CNN-based object detection at the rate of 33 fps with acceleration using graphic processing units (GPUs). Controlling the pan-tilt tracking system via visual feedback at 500 Hz, fast moving objects can be robustly tracked at the center of the camera view. The effectiveness of our method was experimentally demonstrated via several results when fast-moving pre-learned objects, such as toy cars were tracked in complex backgrounds.

中文翻译:

具有基于深度学习的目标检测功能的 500-fps 云台跟踪系统

在这封信中,我们提出了一种快速镜像驱动的云台目标跟踪系统,可以稳健地跟踪外观在复杂背景中以 500 fps 变化的对象。通过假设帧之间的小图像位移,这是高帧率视觉的一个特性,我们通过将基于卷积神经网络 (CNN) 的对象检测与基于模板匹配 (TM) 的跟踪混合,开发了一种快速对象跟踪算法以每秒数百帧 (fps) 的速度运行。对于具有高速视觉反馈的对象跟踪,所提出的跟踪算法可以通过在几毫秒内同时对连续帧的多幅图像执行基于 TM 的跟踪,显着减少基于 CNN 的对象检测中的数十毫秒延迟。在提议的云台跟踪系统中,当当前跟踪的对象被遮挡或不在相机视图中时,它可以使用图形处理单元 (GPU) 以 33 fps 的速度通过基于 CNN 的对象检测来识别要新跟踪的对象。通过 500 Hz 的视觉反馈控制云台跟踪系统,可以在摄像机视图的中心可靠地跟踪快速移动的物体。当在复杂背景中跟踪快速移动的预学习对象(例如玩具车)时,我们的方法的有效性通过几个结果得到了实验证明。可以在相机视图的中心稳健地跟踪快速移动的物体。当在复杂背景中跟踪快速移动的预学习对象(例如玩具车)时,我们的方法的有效性通过几个结果得到了实验证明。可以在相机视图的中心稳健地跟踪快速移动的物体。当在复杂背景中跟踪快速移动的预学习对象(例如玩具车)时,我们的方法的有效性通过几个结果得到了实验证明。
更新日期:2021-04-01
down
wechat
bug