首页> 外文会议>IEEE Conference on Computer Vision and Pattern Recognition >Training Region-Based Object Detectors with Online Hard Example Mining
【24h】

Training Region-Based Object Detectors with Online Hard Example Mining

机译:通过在线硬示例挖掘训练基于区域的对象检测器

获取原文
获取外文期刊封面目录资料

摘要

The field of object detection has made significant advances riding on the wave of region-based ConvNets, but their training procedure still includes many heuristics and hyperparameters that are costly to tune. We present a simple yet surprisingly effective online hard example mining (OHEM) algorithm for training region-based ConvNet detectors. Our motivation is the same as it has always been - detection datasets contain an overwhelming number of easy examples and a small number of hard examples. Automatic selection of these hard examples can make training more effective and efficient. OHEM is a simple and intuitive algorithm that eliminates several heuristics and hyperparameters in common use. But more importantly, it yields consistent and significant boosts in detection performance on benchmarks like PASCAL VOC 2007 and 2012. Its effectiveness increases as datasets become larger and more difficult, as demonstrated by the results on the MS COCO dataset. Moreover, combined with complementary advances in the field, OHEM leads to state-of-the-art results of 78.9% and 76.3% mAP on PASCAL VOC 2007 and 2012 respectively.
机译:在基于区域的ConvNets的浪潮中,对象检测领域取得了长足的进步,但是它们的训练过程仍然包括许多启发式方法和超参数,它们的调整成本很高。我们提出了一种简单但令人惊讶的有效在线硬示例挖掘(OHEM)算法,用于训练基于区域的ConvNet检测器。我们的动机与往常一样-检测数据集包含大量的简单示例和少量的困难示例。自动选择这些困难的例子可以使培训更加有效。 OHEM是一种简单而直观的算法,它消除了一些常用的试探法和超参数。但更重要的是,它可以在PASCAL VOC 2007和2012等基准测试上始终如一地显着提高检测性能。随着MS COCO数据集上的结果证明,其有效性随着数据集的增大和难度的增加而提高。此外,结合该领域的互补进展,OHEM在PASCAL VOC 2007和2012上分别获得了78.9%和76.3%的mAP的最新结果。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号