有对象和无对象的物体识别

Nov, 2016

Object Recognition with and without Objects

Zhuotun Zhu, Lingxi Xie, Alan L. Yuille

TL;DR本文提出了一种训练深度神经网络的新方法，分别对图像前景（对象）和背景（上下文）区域进行训练，并将多个训练网络结合在一起，以提高图像识别性能，证明了所提出的框架的优点。

Abstract

While recent deep neural network models have given promising performance on object recognition, they rely implicitly on the visual contents of the whole image. In this paper, we train deep neural networks on the