arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.13136cs.CV
收藏

少即是多:大空间图像中的高效目标检测

Fewer is More: Efficient Object Detection in Large Aerial Images

Northwestern Polytechnical University(西北工业大学);Zhengzhou Institute of Surveying and Mapping(郑州测绘学院)

Xingxing Xie, Gong Cheng, Qingyang Li, Shicheng Miao, Ke Li, Junwei Han

摘要:

目前主流的大幅面航空影像目标检测目标检测方法通常是将大幅面航空影像划分为小块,然后在所有小块上穷尽检测感兴趣的目标,而不管是否存在目标。这种模式虽然有效,但效率低下,因为检测器必须通过所有补丁,严重阻碍了推理速度。本文提出了一种对象激活网络(OAN),以帮助检测器专注于更少的块,但实现更有效的推理和更准确的结果,从而实现一种简单有效的解决方案,大图像中的目标检测。简单地说,OAN是一个轻量级的全卷积网络,用于判断每个补丁是否包含对象,它可以很容易地集成到许多对象检测器中,并与它们端到端地联合训练。我们用五个先进的探测器广泛评估我们的OAN。使用OAN,所有五个检测器在三个大规模航空影像数据集上获得超过30.0%的速度提升,同时具有一致的精度提高。在极大的Gaofen-2图像(29200 $\times$ 27620像素),我们的OAN提高了70.5%的检测速度。此外,我们将OAN扩展到驾驶场景目标检测和4K视频目标检测,分别将检测速度提高了112.1%和75.0%,而不牺牲准确性。代码可在https://github.com/Ranchosky/OAN上获得。

英文摘要:

Current mainstream object detection methods for large aerial images usually divide large images into patches and then exhaustively detect the objects of interest on all patches, no matter whether there exist objects or not. This paradigm, although effective, is inefficient because the detectors have to go through all patches, severely hindering the inference speed. This paper presents an Objectness Activation Network (OAN) to help detectors focus on fewer patches but achieve more efficient inference and more accurate results, enabling a simple and effective solution to object detection in large images. In brief, OAN is a light fully-convolutional network for judging whether each patch contains objects or not, which can be easily integrated into many object detectors and jointly trained with them end-to-end. We extensively evaluate our OAN with five advanced detectors. Using OAN, all five detectors acquire more than 30.0% speed-up on three large-scale aerial image datasets, meanwhile with consistent accuracy improvements. On extremely large Gaofen-2 images (29200$\times$27620 pixels), our OAN improves the detection speed by 70.5%. Moreover, we extend our OAN to driving-scene object detection and 4K video object detection, boosting the detection speed by 112.1% and 75.0%, respectively, without sacrificing the accuracy. Code is available at https://github.com/Ranchosky/OAN.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑