DL之FastR-CNN：Fast R-CNN算法的简介(论文介绍)、架构详解、案例应用等配图集合之详细攻略 / 四六文摘

DL之FastR-CNN：Fast R-CNN算法的简介(论文介绍)、架构详解、案例应用等配图集合之详细攻略相关文章DL之R-CNN：R-CNN算法的简介(论文介绍)、架构详解、案例应用等配图集合之详细攻略DL之FastR-CNN：Fast R-CNN算法的简介(论文介绍)、架构详解、案例应用等配图集合之详细攻略DL之FastR-CNN：Fast R-CNN算法的架构详解DL之FasterR-CNN：Faster R-CNN算法的简介(论文介绍)、架构详解、案例应用等配图集合之详细攻略Fast R-CNN算法的简介(论文介绍)AbstractThis paper proposes a Fast Region-based Convolutional Network method (Fast R-CNN) for object detection. Fast R-CNN builds on previous work to efficiently classify object proposals using deep convolutional networks. Compared to previous work, Fast R-CNN employs several innovations to improve training and testing speed while also increasing detection accuracy. Fast R-CNN trains the very deep VGG16 network 9× faster than R-CNN, is 213× faster at test-time, and achieves a higher mAP on PASCAL VOC 2012. Compared to SPPnet, Fast R-CNN trains VGG16 3× faster, tests 10× faster, and is more accurate. Fast R-CNN is implemented in Python and C++ (using Caffe) and is available under the open-source MIT License at https: //github.com/rbgirshick/fast-rcnn.摘要本文提出了一种基于区域卷积网络的快速目标检测方法(Fast R-CNN)。Fast R-CNN建立在以前工作的基础上，使用深度卷积网络有效地分类对象建议。与之前的工作相比，Fast R-CNN在提高训练和测试速度的同时，也提高了检测精度。Fast R-CNN训练了非常深的VGG16网络，速度比R-CNN快9倍，测试时速度213倍，在PASCAL VOC 2012上实现了更高的mAP。与SPPnet相比，Fast R-CNN训练VGG16 更快3倍，测试速度更快10倍，更准确。Fast R-CNN是用Python和c++(使用Caffe)实现的，可以通过https: //github.com/rbgirshick/fast-rcnn获得MIT的开源许可。ConclusionThis paper proposes Fast R-CNN, a clean and fast update to R-CNN and SPPnet. In addition to reporting state-of-theart detection results, we present detailed experiments that we hope provide new insights. Of particular note, sparse object proposals appear to improve detector quality. This issue was too costly (in time) to probe in the past, but becomes practical with Fast R-CNN. Of course, there may exist yet undiscovered techniques that allow dense boxes to perform as well as sparse proposals. Such methods, if developed, may help further accelerate object detection.结论本文提出了一种Fast R-CNN，对R-CNN和SPPnet进行了干净、快速的更新。除了报告最新的检测结果，我们还提供了详细的实验，希望能提供新的见解。特别值得注意的是，稀疏对象建议似乎可以提高检测器的质量。这个问题在过去花费太多(时间)去探索，但在Fast R-CNN中变得实用。当然，可能存在一些尚未发现的技术，允许密集的框执行稀疏的建议。如果开发出这样的方法，将有助于进一步加速目标检测。论文Ross Girshick.Fast R-CNN. ICCV 2015https://arxiv.org/abs/1504.080831、实验结果1、mAPbased on PASCAL VOC 2007, results from Girshick图示可知，SPP比R-CNN的测试速度快了24倍，而Fast R-CNN又比SPP加快了10倍以上！