ScaleNet - Improve CNNs through Recursively Rescaling Objects
ScaleNet - Improve CNNs through Recursively Rescaling Objects
复制标题
DOI：
10.1609/aaai.v34i07.6806
复制
发表时间：
2020-04
期刊：
影响因子：
--
通讯作者：
Xingyi Li;Zhongang Qi;Xiaoli Z. Fern;Li Fuxin
中科院分区：
文献类型：
--
作者：
Xingyi Li;Zhongang Qi;Xiaoli Z. Fern;Li Fuxin
深度网络通常不是尺度不变的，因此如果可识别的对象处于仅在测试时发生的不可见尺度，则其性能可能会有很大的变化。在本文中，我们提出了ScaleNet，它在深度学习框架中递归预测对象规模。ScaleNet的明确目标是预测图像中对象的尺度，它使预训练的深度学习模型能够识别其训练集中不存在的尺度中的对象。通过递归调用ScaleNet，可以推广到训练集中看不到的非常大的规模变化。为了证明我们提出的框架的鲁棒性，我们在MNIST，CIFAR-10和MS COCO数据集上使用预训练和微调的分类和检测框架进行了实验，结果表明我们提出的框架显着提高了深度网络的性能。
Deep networks are often not scale-invariant hence their performance can vary wildly if recognizable objects are at an unseen scale occurring only at testing time. In this paper, we propose ScaleNet, which recursively predicts object scale in a deep learning framework. With an explicit objective to predict the scale of objects in images, ScaleNet enables pretrained deep learning models to identify objects in the scales that are not present in their training sets. By recursively calling ScaleNet, one can generalize to very large scale changes unseen in the training set. To demonstrate the robustness of our proposed framework, we conduct experiments with pretrained as well as fine-tuned classification and detection frameworks on MNIST, CIFAR-10, and MS COCO datasets and results reveal that our proposed framework significantly boosts the performances of deep networks.