{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2023,11,25]],"date-time":"2023-11-25T07:02:42Z","timestamp":1700895762423},"reference-count":49,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2021,10,26]],"date-time":"2021-10-26T00:00:00Z","timestamp":1635206400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurorobot."],"abstract":"Generative adversarial networks and variational autoencoders (VAEs) provide impressive image generation from Gaussian white noise, but both are difficult to train, since they need a generator (or encoder) and a discriminator (or decoder) to be trained simultaneously, which can easily lead to unstable training. To solve or alleviate these synchronous training problems of generative adversarial networks (GANs) and VAEs, researchers recently proposed generative scattering networks (GSNs), which use wavelet scattering networks (ScatNets) as the encoder to obtain features (or ScatNet embeddings) and convolutional neural networks (CNNs) as the decoder to generate an image. The advantage of GSNs is that the parameters of ScatNets do not need to be learned, while the disadvantage of GSNs is that their ability to obtain representations of ScatNets is slightly weaker than that of CNNs. In addition, the dimensionality reduction method of principal component analysis (PCA) can easily lead to overfitting in the training of GSNs and, therefore, affect the quality of generated images in the testing process. To further improve the quality of generated images while keeping the advantages of GSNs, this study proposes generative fractional scattering networks (GFRSNs), which use more expressive fractional wavelet scattering networks (FrScatNets), instead of ScatNets as the encoder to obtain features (or FrScatNet embeddings) and use similar CNNs of GSNs as the decoder to generate an image. Additionally, this study develops a new dimensionality reduction method named feature-map fusion (FMF) instead of performing PCA to better retain the information of FrScatNets,; it also discusses the effect of image fusion on the quality of the generated image. The experimental results obtained on the CIFAR-10 and CelebA datasets show that the proposed GFRSNs can lead to better generated images than the original GSNs on testing datasets. The experimental results of the proposed GFRSNs with deep convolutional GAN (DCGAN), progressive GAN (PGAN), and CycleGAN are also given.<\/jats:p>","DOI":"10.3389\/fnbot.2021.752752","type":"journal-article","created":{"date-parts":[[2021,10,26]],"date-time":"2021-10-26T06:51:01Z","timestamp":1635231061000},"update-policy":"http:\/\/dx.doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Fractional Wavelet-Based Generative Scattering Networks"],"prefix":"10.3389","volume":"15","author":[{"given":"Jiasong","family":"Wu","sequence":"first","affiliation":[]},{"given":"Xiang","family":"Qiu","sequence":"additional","affiliation":[]},{"given":"Jing","family":"Zhang","sequence":"additional","affiliation":[]},{"given":"Fuzhi","family":"Wu","sequence":"additional","affiliation":[]},{"given":"Youyong","family":"Kong","sequence":"additional","affiliation":[]},{"given":"Guanyu","family":"Yang","sequence":"additional","affiliation":[]},{"given":"Lotfi","family":"Senhadji","sequence":"additional","affiliation":[]},{"given":"Huazhong","family":"Shu","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2021,10,26]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"433","DOI":"10.1002\/wics.101","article-title":"Principal component analysis","volume":"2","author":"Abdi","year":"2010","journal-title":"Wiley Interdisc. Rev. Comput. Stat"},{"key":"B2","article-title":"Generative networks as inverse problems with scattering transforms","volume-title":"2018 International Conference on Learning Representations (ICLR)","author":"Angles","year":"2018"},{"key":"B3","first-page":"214","article-title":"Wasserstein generative adversarial networks","volume-title":"2017 International Conference on Machine Learning (ICML)","author":"Arjovsky","year":"2017"},{"key":"B4","doi-asserted-by":"publisher","first-page":"859","DOI":"10.1080\/01621459.2017.1285773","article-title":"Variational inference: a review for statisticians","volume":"112","author":"Blei","year":"2017","journal-title":"J. Am. Stat. Assoc."},{"key":"B5","article-title":"Large scale GAN training for high fidelity natural image synthesis","volume-title":"2019 International Conference on Learning Representations (ICLR)","author":"Brock","year":"2019"},{"key":"B6","doi-asserted-by":"publisher","first-page":"1872","DOI":"10.1109\/TPAMI.2012.230","article-title":"Invariant scattering convolution networks","volume":"35","author":"Bruna","year":"2013","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell"},{"key":"B7","first-page":"2180","article-title":"InfoGAN: interpretable representation learning by information maximizing generative adversarial nets","volume-title":"2016 Proceedings of the 30th International Conference on Neural Information Processing Systems, (NeurIPS)","author":"Chen","year":"2016"},{"key":"B8","first-page":"2414","article-title":"Image style transfer using convolutional neural networks","volume-title":"IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Gatys","year":"2016"},{"key":"B9","doi-asserted-by":"publisher","first-page":"2672","DOI":"10.3156\/JSOFT.29.5_177_2","article-title":"Generative adversarial nets","volume":"27","author":"Goodfellow","year":"2014","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B10","article-title":"Temporal difference variational auto-encoder","volume-title":"2018 International Conference on Learning Representations (ICLR)","author":"Gregor","year":"2018"},{"key":"B11","first-page":"5769","article-title":"Improved training of wasserstein GANs","volume-title":"2017 Proceedings of the 31st International Conference on Neural Information Processing Systems (NeurIPS)","author":"Gulrajani","year":"2017"},{"key":"B12","article-title":"beta-VAE: learning basic visual concepts with a constrained variational framework","volume-title":"2017 International Conference on Learning Representations (ICLR)","author":"Higgins","year":"2017"},{"key":"B13","first-page":"4487","article-title":"Generative adversarial transformers","volume-title":"Proceedings of the 38th International Conference on Machine Learning (ICML)","author":"Hudson","year":"2021"},{"key":"B14","first-page":"448","article-title":"Batch normalization: accelerating deep network training by reducing internal covariate shift","volume-title":"Proceedings of The 32nd International Conference on Machine Learning (ICML)","author":"Ioffe","year":"2015"},{"key":"B15","article-title":"Progressive growing of GANs for improved quality, stability, and variation","volume-title":"2018 International Conference on Learning Representations","author":"Karras","year":"2018"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TPAMI.2020.2970919","article-title":"A style-based generator architecture for generative adversarial networks","volume":"1","author":"Karras","year":"","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"B17","first-page":"8110","article-title":"Analyzing and improving the image quality of StyleGAN","volume-title":"2020 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Karras","year":""},{"key":"B18","article-title":"Adam: a method for stochastic optimization","volume-title":"2015 International Conference on Learning Representations (ICML)","author":"Kingma","year":"2015"},{"key":"B19","article-title":"Auto-encoding variational Bayes","volume-title":"2014 International Conference on Learning Representations (ICLR)","author":"Kingma","year":"2014"},{"key":"B20","author":"Krizhevsky","year":"2009"},{"key":"B21","doi-asserted-by":"publisher","first-page":"553","DOI":"10.1109\/TBME.2018.2850356","article-title":"Fractional wavelet scattering network and applications","volume":"66","author":"Liu","year":"2019","journal-title":"IEEE Trans. Biomed. Eng"},{"key":"B22","doi-asserted-by":"publisher","first-page":"1882","DOI":"10.1109\/LSP.2016.2618776","article-title":"Image fusion with convolutional sparse representation","volume":"23","author":"Liu","year":"2016","journal-title":"IEEE Signal Process. Lett"},{"key":"B23","first-page":"3730","author":"Liu","year":"2015"},{"key":"B24","doi-asserted-by":"crossref","first-page":"3431","DOI":"10.1109\/CVPR.2015.7298965","article-title":"Fully convolutional networks for semantic segmentation","volume-title":"2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Long","year":"2015"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1331","DOI":"10.1002\/cpa.21413","article-title":"Group invariant scattering","volume":"65","author":"Mallat","year":"2012","journal-title":"Commun. Pure Appl. Math"},{"key":"B26","article-title":"Conditional generative adversarial nets","volume-title":"arXiv","author":"Mirza","year":"2014"},{"key":"B27","article-title":"Spectral normalization for generative adversarial networks","volume-title":"2018 International Conference on Learning Representations (ICLR)","author":"Miyato","year":"2018"},{"key":"B28","first-page":"807","article-title":"Rectified linear units improve restricted boltzmann machines","volume-title":"Proceedings of the 27th International Conference on Machine Learning (ICML)","author":"Nair","year":"2010"},{"key":"B29","first-page":"271","article-title":"f -GAN: training generative neural samplers using variational divergence minimization","volume-title":"2016 Proceedings of the 30th International Conference on Neural Information Processing Systems (NeurIPS)","author":"Nowozin","year":"2016"},{"key":"B30","first-page":"6306","article-title":"Neural discrete representation learning","volume-title":"2017 Proceedings of the 31st International Conference on Neural Information Processing Systems (NeurIPS)","author":"Oord","year":"2017"},{"key":"B31","doi-asserted-by":"publisher","first-page":"1118","DOI":"10.1007\/s11263-019-01265-2","article-title":"Loss-sensitive generative adversarial networks on lipschitz densities","volume":"128","author":"Qi","year":"2020","journal-title":"Int. J. Comput. Vis."},{"key":"B32","article-title":"Unsupervised representation learning with deep convolutional generative adversarial networks","volume-title":"2016 International Conference on Learning Representations (ICLR)","author":"Radford","year":"2016"},{"key":"B33","first-page":"14837","article-title":"Generating diverse high-fidelity images with VQ-VAE-2","volume-title":"2019 Proceedings of the 33rd International Conference on Neural Information Processing Systems (NeurIPS)","author":"Razavi","year":"2019"},{"key":"B34","first-page":"1530","article-title":"Variational inference with normalizing flows","volume-title":"2015 Proceedings of the 32nd International Conference on Machine Learning (ICML)","author":"Rezende","year":"2015"},{"key":"B35","first-page":"1278","article-title":"Stochastic backpropagation and approximate inference in deep generative models","volume-title":"2014 Proceedings of the 31st International Conference on Machine Learning (ICML)","author":"Rezende","year":"2014"},{"key":"B36","article-title":"Progressive neural networks","volume-title":"arXiv","author":"Rusu","year":"2017"},{"key":"B37","first-page":"1218","article-title":"Markov Chain Monte Carlo and variational inference: bridging the gap","volume-title":"Proceedings of the 32nd International Conference on Machine Learning (ICML)","author":"Salimans","year":"2015"},{"key":"B38","doi-asserted-by":"publisher","first-page":"909","DOI":"10.1109\/LSP.2010.2071383","article-title":"Multichannel sampling and reconstruction of bandlimited signals in fractional fourier domain","volume":"17","author":"Shi","year":"2010","journal-title":"IEEE Signal Process. Lett"},{"key":"B39","first-page":"412","article-title":"GraphVAE: towards generation of small graphs using variational autoencoders","volume-title":"27th International Conference on Artificial Neural Networks (ICANN)","author":"Simonovsky","year":"2018"},{"key":"B40","first-page":"3483","article-title":"Learning structured output representation using deep conditional generative models","volume-title":"2015 Proceedings of the 29th International Conference on Neural Information Processing Systems, (NeurIPS)","author":"Sohn","year":"2015"},{"key":"B41","first-page":"3738","article-title":"Ladder variational autoencoders","volume-title":"2016 30th Annual Conference on Neural Information Processing Systems Conference (NeurIPS)","author":"S\u00f8nderby","year":"2016"},{"key":"B42","article-title":"Differentially private learning needs better features (or much more data)","volume-title":"2021 International Conference on Learning Representations (ICLR)","author":"Tramer","year":"2021"},{"key":"B43","doi-asserted-by":"crossref","first-page":"994","DOI":"10.1109\/ICNNSP.2003.1281036","article-title":"Image quality evaluation based on image weighted separating block peak signal to noise ratio","volume-title":"Proceedings of the 2003 International Conference on Neural Networks and Signal Processing","author":"Wang","year":"2003"},{"key":"B44","doi-asserted-by":"publisher","first-page":"600","DOI":"10.1109\/TIP.2003.819861","article-title":"Image quality assessment: from error visibility to structural similarity","volume":"13","author":"Wang","year":"2004","journal-title":"IEEE Trans. Image Proc."},{"key":"B45","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3459992","article-title":"Generative adversarial networks in computer vision: a survey and taxonomy","volume":"54","author":"Wang","year":"2021","journal-title":"ACM Comput. Surveys"},{"key":"B46","doi-asserted-by":"publisher","first-page":"1750037","DOI":"10.1142\/S0219691317500370","article-title":"Multi-focus image fusion and super-resolution with convolutional neural network","volume":"15","author":"Yang","year":"2017","journal-title":"Int. J. Wavelets Multiresol. Inform. Proc"},{"key":"B47","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1109\/TCI.2016.2644865","article-title":"Loss functions for image restorationwith neural networks","volume":"3","author":"Zhao","year":"2017","journal-title":"IEEE Transac. Comput. Imaging"},{"key":"B48","first-page":"2242","article-title":"Unpaired image-to-image translation using cycle-consistent adversarial networks","author":"Zhu","year":"2017","journal-title":"2017 IEEE International Conference on Computer Vision"},{"key":"B49","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/IJCNN.2019.8851705","article-title":"Encoding robust representation for graph generation","volume-title":"2019 International Joint Conference on Neural Networks (IJCNN)","author":"Zou","year":"2019"}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2021.752752\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,10,26]],"date-time":"2021-10-26T06:51:29Z","timestamp":1635231089000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2021.752752\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,10,26]]},"references-count":49,"alternative-id":["10.3389\/fnbot.2021.752752"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2021.752752","relation":{},"ISSN":["1662-5218"],"issn-type":[{"value":"1662-5218","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,10,26]]}}}
