Quaternion Convolutional Neural Networks

Neural networks in the real domain have been studied for a long time and achieved promising results in many vision tasks for recent years. However, the extensions of the neural network models in other number fields and their potential applications are not fully-investigated yet. Focusing on color images, which can be naturally represented as quaternion matrices, we propose a quaternion convolutional neural network (QCNN) model to obtain more representative features. In particular, we re-design the basic modules like convolution layer and fully-connected layer in the quaternion domain, which can be used to establish fully-quaternion convolutional neural networks. Moreover, these modules are compatible with almost all deep learning techniques and can be plugged into traditional CNNs easily. We test our QCNN models in both color image classification and denoising tasks. Experimental results show that they outperform the real-valued CNNs with same structures.

[1]  Li Fei-Fei,et al.  ImageNet: A large-scale hierarchical image database , 2009, CVPR.

[2]  Licheng Yu,et al.  Vector Sparse Representation of Color Image Using Quaternion Matrix Analysis , 2015, IEEE Transactions on Image Processing.

[3]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Akira Hirose,et al.  Generalization Characteristics of Complex-Valued Feedforward Neural Networks in Relation to Signal Coherence , 2012, IEEE Transactions on Neural Networks and Learning Systems.

[5]  Licheng Yu,et al.  Quaternion-based sparse representation of color image , 2013, 2013 IEEE International Conference on Multimedia and Expo (ICME).

[6]  Kaiming He,et al.  Mask R-CNN , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[7]  Guowei Yang,et al.  Sparse representation based on vector extension of reduced quaternion matrix for multiscale image denoising , 2016, IET Image Process..

[8]  Jiasong Wu,et al.  Color image classification via quaternion principal component analysis network , 2015, Neurocomputing.

[9]  S. Sangwine.  Fourier transforms of colour images using quaternion or hypercomplex, numbers , 1996 .

[10]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[11]  T. Nitta,et al.  A quaternary version of the back-propagation algorithm , 1995, Proceedings of ICNN'95 - International Conference on Neural Networks.

[12]  Andrew Zisserman,et al.  Return of the Devil in the Details: Delving Deep into Convolutional Nets , 2014, BMVC.

[13]  Enhong Chen,et al.  Image Denoising and Inpainting with Deep Neural Networks , 2012, NIPS.

[14]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[15]  Lawrence D. Jackel,et al.  Backpropagation Applied to Handwritten Zip Code Recognition , 1989, Neural Computation.

[16]  Tohru Nitta,et al.  On the critical points of the complex-valued neural network , 2002, Proceedings of the 9th International Conference on Neural Information Processing, 2002. ICONIP '02..

[17]  Yoshua Bengio,et al.  Extracting and composing robust features with denoising autoencoders , 2008, ICML '08.

[18]  Yaser Sheikh,et al.  OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[19]  Xiaokang Yang,et al.  QWT: Retrospective and New Applications , 2010, Geometric Algebra Computing.

[20]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[21]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[22]  Alex Krizhevsky,et al.  Learning Multiple Layers of Features from Tiny Images , 2009 .

[23]  Xiaoou Tang,et al.  Learning a Deep Convolutional Network for Image Super-Resolution , 2014, ECCV.

[24]  Eduardo Bayro-Corrochano,et al.  The Motor Extended Kalman Filter: A Geometric Approach for Rigid Motion Estimation , 2004, Journal of Mathematical Imaging and Vision.

[25]  Eduardo Bayro-Corrochano,et al.  Geometric neural computing , 2001, IEEE Trans. Neural Networks.

[26]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[27]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[28]  Creed F. Jones.  Color Face Recognition using Quaternionic Gabor Filters , 2005 .

[29]  Sandeep Subramanian,et al.  Deep Complex Networks , 2017, ICLR.

[30]  Anthony S. Maida,et al.  Deep Quaternion Networks , 2017, 2018 International Joint Conference on Neural Networks (IJCNN).

[31]  P. Girard,et al.  The quaternion group and modern physics , 1984 .

[32]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[33]  Andrew Zisserman,et al.  Automated Flower Classification over a Large Number of Classes , 2008, 2008 Sixth Indian Conference on Computer Vision, Graphics & Image Processing.

[34]  Eduardo Bayro-Corrochano,et al.  The Theory and Use of the Quaternion Wavelet Transform , 2005, Journal of Mathematical Imaging and Vision.

[35]  Akira Hirose,et al.  Quaternion Neural-Network-Based PolSAR Land Classification in Poincare-Sphere-Parameter Space , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[36]  Pietro Perona,et al.  Microsoft COCO: Common Objects in Context , 2014, ECCV.

[37]  Yi Xu,et al.  Single color image super-resolution using quaternion-based sparse representation , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[38]  Yu-Bin Yang,et al.  Image Restoration Using Very Deep Convolutional Encoder-Decoder Networks with Symmetric Skip Connections , 2016, NIPS.

[39]  Robert B. McGhee,et al.  Design, Implementation, and Experimental Results of a Quaternion-Based Kalman Filter for Human Body Motion Tracking , 2005, IEEE Transactions on Robotics.