作者：Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun
发布时间：2014
发布期刊：ECCV
论文全称：Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
论文地址：https://arxiv.org/pdf/1406.4729.pdf
代码：*http://research.microsoft.com/en-us/um/people/kahe/*
地位：提出了空间金字塔池化
Existing deep convolutional neural networks (CNNs) require a fifixed-size (e.g., 224×224) input image. This requirement is “artifificial” and may reduce the recognition accuracy for the images or sub-images of an arbitrary size/scale. In this work, we equip the networks with another pooling strategy, “spatial pyramid pooling”, to eliminate the above requirement. The new network structure, called SPP-net, can generate a fifixed-length representation regardless of image size/scale. Pyramid pooling is also robust to object deformations. With these advantages, SPP-net should in general improve all CNN-based image classifification methods. On the ImageNet 2012 dataset, we demonstrate that SPP-net boosts the accuracy of a variety of CNN architectures despite their different designs. On the Pascal VOC 2007 and Caltech101 datasets, SPP-net achieves state-of-the art classifification results using a single full-image representation and no fifine-tuning.
The power of SPP-net is also signifificant in object detection. Using SPP-net, we compute the feature maps from the entire image only once, and then pool features in arbitrary regions (sub-images) to generate fifixed-length representations for training the detectors. This method avoids repeatedly computing the convolutional features. In processing test images, our method is 24-102× faster than the R-CNN method, while achieving better or comparable accuracy on Pascal VOC 2007.
In ImageNet Large Scale Visual Recognition Challenge (ILSVRC) 2014, our methods rank #2 in object detection and #3 in image classification among all 38 teams. This manuscript also introduces the improvement made for this competition.
普遍的cnn需要固定的输入图像大小，这样的缺点有：
即固定输入大小忽略了图像比例的问题