Geometric Data Augmentation Based on Feature Map Ensemble

  title={Geometric Data Augmentation Based on Feature Map Ensemble},
  author={Takashi Shibata and Masayuki Tanaka and M. Okutomi},
Deep convolutional networks have become the mainstream in computer vision applications. Although CNNs have been successful in many computer vision tasks, it is not free from drawbacks. The performance of CNN is dramatically degraded by geometric transformation, such as large rotations. In this paper, we propose a novel CNN architecture that can improve the robustness against geometric transformations without modifying the existing backbones of their CNNs. The key is to enclose the existing… Expand

Figures and Tables from this paper


Towards Learning Affine-Invariant Representations via Data-Efficient CNNs
A novel multi-scale maxout CNN is proposed and train it end-to-end with a novel rotation-invariant regularizer that aims to enforce the weights in each 2D spatial filter to approximate circular patterns. Expand
MINTIN: Maxout-Based and Input-Normalized Transformation Invariant Neural Network
A Maxout-based and input-normalized transformation invariant neural network (MINTIN), which aims at addressing the nuisance variation of images and accumulating transformation invariance, and introduces an innovative module, the Normalization, and combine it with the Maxout operator. Expand
Some Improvements on Deep Convolutional Neural Network Based Image Classification
This paper summarizes the entry in the Imagenet Large Scale Visual Recognition Challenge 2013, which achieved a top 5 classification error rate and achieved over a 20% relative improvement on the previous year's winner. Expand
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
This work introduces a Region Proposal Network (RPN) that shares full-image convolutional features with the detection network, thus enabling nearly cost-free region proposals and further merge RPN and Fast R-CNN into a single network by sharing their convolutionAL features. Expand
Spatial Transformer Networks
This work introduces a new learnable module, the Spatial Transformer, which explicitly allows the spatial manipulation of data within the network, and can be inserted into existing convolutional architectures, giving neural networks the ability to actively spatially transform feature maps. Expand
Harmonic Networks: Deep Translation and Rotation Equivariance
H-Nets are presented, a CNN exhibiting equivariance to patch-wise translation and 360-rotation, and it is demonstrated that their layers are general enough to be used in conjunction with the latest architectures and techniques, such as deep supervision and batch normalization. Expand
Squeeze-and-Excitation Networks
This work proposes a novel architectural unit, which is term the “Squeeze-and-Excitation” (SE) block, that adaptively recalibrates channel-wise feature responses by explicitly modelling interdependencies between channels and shows that these blocks can be stacked together to form SENet architectures that generalise extremely effectively across different datasets. Expand
TI-POOLING: Transformation-Invariant Pooling for Feature Learning in Convolutional Neural Networks
A deep neural network topology that incorporates a simple to implement transformationinvariant pooling operator (TI-POOLING) that is able to efficiently handle prior knowledge on nuisance variations in the data, such as rotation or scale changes is presented. Expand
Fast AutoAugment
This paper proposes an algorithm called Fast AutoAugment that finds effective augmentation policies via a more efficient search strategy based on density matching that speeds up the search time by orders of magnitude while achieves comparable performances on image recognition tasks with various models and datasets. Expand
Densely Connected Convolutional Networks
The Dense Convolutional Network (DenseNet), which connects each layer to every other layer in a feed-forward fashion, and has several compelling advantages: they alleviate the vanishing-gradient problem, strengthen feature propagation, encourage feature reuse, and substantially reduce the number of parameters. Expand