TI-POOLING: Transformation-Invariant Pooling for Feature Learning in Convolutional Neural Networks
2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)(2016)
摘要
In this paper we present a deep neural network topology that incorporates a simple to implement transformationinvariant pooling operator (TI-POOLING). This operator is able to efficiently handle prior knowledge on nuisance variations in the data, such as rotation or scale changes. Most current methods usually make use of dataset augmentation to address this issue, but this requires larger number of model parameters and more training data, and results in significantly increased training time and larger chance of under-or overfitting. The main reason for these drawbacks is that that the learned model needs to capture adequate features for all the possible transformations of the input. On the other hand, we formulate features in convolutional neural networks to be transformation-invariant. We achieve that using parallel siamese architectures for the considered transformation set and applying the TI-POOLING operator on their outputs before the fully-connected layers. We show that this topology internally finds the most optimal "canonical" instance of the input image for training and therefore limits the redundancy in learned features. This more efficient use of training data results in better performance on popular benchmark datasets with smaller number of parameters when comparing to standard convolutional neural networks with dataset augmentation and to other baselines.
更多查看译文
关键词
TI-POOLING,transformation-invariant pooling,feature learning,convolutional neural networks,deep neural network topology,nuisance data variations,dataset augmentation,parallel siamese architectures,optimal canonical instance,training data
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络