arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2010.15823cs.CVcs.LG
收藏

目标探测器尺度的黑盒优化

Black-Box Optimization of Object Detector Scales

Hochschule Bonn-Rhein-Sieg(波恩-莱茵-锡格应用科学大学);University of Bremen(不来梅大学);German Research Center for Artificial Intelligence(德国人工智能研究中心)

Mohandass Muthuraja, Octavio Arriaga, Paul Plöger, Frank Kirchner, Matias Valdenegro-Toro

摘要:

在过去的几年中,通过使用先进的CNN架构,对象检测器已经有了很大的改进。然而,许多检测器超参数通常是手动调整的,或者它们与检测器作者设置的值一起使用。自动超参数优化尚未在改进基于CNN的对象检测器超参数中进行探索。在这项工作中,我们建议使用黑盒优化方法来调整更快的R-CNN和SSD中的先验/默认框尺度,使用贝叶斯优化,SMAC和CMA-ES。我们表明,通过调整输入图像大小和Faster R-CNN上的先前框锚标度,mAP在PASCAL VOC 2007上增加了2%,在SSD上增加了3%。在使用SSD的COCO数据集上,中型和大型对象的mAP有所改善,但小型对象的mAP下降了1%。我们还进行了回归分析,以找到要调整的重要超参数。

英文摘要:

Object detectors have improved considerably in the last years by using advanced CNN architectures. However, many detector hyper-parameters are generally manually tuned, or they are used with values set by the detector authors. Automatic Hyper-parameter optimization has not been explored in improving CNN-based object detectors hyper-parameters. In this work, we propose the use of Black-box optimization methods to tune the prior/default box scales in Faster R-CNN and SSD, using Bayesian Optimization, SMAC, and CMA-ES. We show that by tuning the input image size and prior box anchor scale on Faster R-CNN mAP increases by 2% on PASCAL VOC 2007, and by 3% with SSD. On the COCO dataset with SSD there are mAP improvement in the medium and large objects, but mAP decreases by 1% in small objects. We also perform a regression analysis to find the significant hyper-parameters to tune.

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。

↑