Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition

TL;DR

This paper introduces spatial pyramid pooling in CNNs, enabling fixed-size representations for arbitrary image scales, improving accuracy and efficiency in image classification and object detection tasks.

Contribution

The paper proposes the SPP-net architecture that removes the fixed input size constraint in CNNs, enhancing recognition accuracy and computational efficiency across various visual recognition tasks.

Findings

Boosts CNN accuracy on ImageNet dataset.

Achieves state-of-the-art results on Pascal VOC 2007 and Caltech101.

Significantly speeds up object detection, reducing computation time by 24-102x.

Abstract

Existing deep convolutional neural networks (CNNs) require a fixed-size (e.g., 224x224) input image. This requirement is "artificial" 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 fixed-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 classification 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 classification results using a single full-image representation and no fine-tuning. The power of SPP-net is also significant 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 fixed-length representations for training the detectors. This method avoids repeatedly computing the convolutional features. In processing test images, our method is 24-102x 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.