Desarrollo de un prototipo de software de generación de bocetos miniatura con redes neuronales

dc.contributor.advisorNiño Vasquez, Luis Fernando
dc.contributor.authorDiaz Pinilla, Sergio Alejandro
dc.contributor.researchgroupLABORATORIO DE INVESTIGACIÓN EN SISTEMAS INTELIGENTES - LISIspa
dc.date.accessioned2021-04-06T14:51:29Z
dc.date.available2021-04-06T14:51:29Z
dc.date.issued2020
dc.description.abstractLos bocetos miniatura thumbnails son imágenes sencillas usadas en el área de arte conceptual digital con la finalidad de construir, diseñar o ejecutar una idea, antes de iniciar la producción de la imagen final. Este proceso ayuda a un artista a tener un mejor concepto de las ideas que quiere plasmar, visualizar ideas y descartar elementos que pueden no funcionar. Este proceso permite la iteración rápida de ideas, por lo que es una herramienta muy útil para los artistas en su proceso creativo. El objetivo principal de este trabajo fue desarrollar un prototipo de software que mediante redes neuronales sea capaz de generar bocetos miniatura. Se consultó en la literatura las redes neuronales usadas para la tarea de generación de imágenes, que dominios de imágenes se han usado y los métodos que se usan para evaluar el desempeño. Por consiguiente, mediante el uso de una metodología general basada en aprendizaje de máquina se seleccionó y entrenó una red neuronal con el objetivo de generar estos bocetos. Durante este proceso se construyó un conjunto de datos de bocetos, se seleccionó la red neuronal StyleGAN para el proceso de entrenamiento, también se evaluó si los bocetos generados por la red cumplían con los criterios de calidad usando métricas encontradas en la literatura existente sobre este tema. Finalmente, se implementó un aplicativo que permite generar bocetos desde una página web haciendo uso de la red entrenada.spa
dc.description.abstractThumbnails are simple images used in the area of digital conceptual art in order to build, design or execute an idea, before starting the production of the final image. This process helps the artist to have a better concept of the ideas they want to capture, visualize ideas that may or may not work, and discard elements that may not work. This process allows for the rapid iteration of ideas, making it a very useful tool for artists in their creative process. The main goal of this work was to develop a software prototype that through neural networks is capable of generating miniature sketches. The neural networks used for the imaging task, which image domains have been used, and the methods used to evaluate performance were reviewd from the literature. Therefore, through the use of a general methodology based on machine learning, a neural network type was selected and trained in order to generate such sketches. During this process, we built a set of sketch data, StyleGAN was selected for the training process, also, it was so evaluated whether the sketches generated by the network met the quality criteria using found metrics. Finally, a software application was implemented which allows the generation of sketches through a web page using the trained network.eng
dc.description.degreelevelMaestríaspa
dc.description.methodsMetodología general basada en aprendizaje de máquina se seleccionó y entrenó una red neuronal con el objetivo de generar estos bocetos. Durante este proceso se construyó un conjunto de datos de bocetos, se seleccionó la red neuronal StyleGAN para el proceso de entrenamiento, también se evaluó si los bocetos generados por la red cumplían con los criterios de calidad usando métricas encontradas en la literatura existente sobre este tema. Finalmente, se implementó un aplicativo que permite generar bocetos desde una página web haciendo uso de la red entrenada.spa
dc.description.researchareaSistemas Inteligentesspa
dc.format.extent1 recurso en línea (81 páginas)spa
dc.format.mimetypeapplication/pdfspa
dc.identifier.urihttps://repositorio.unal.edu.co/handle/unal/79379
dc.language.isospaspa
dc.publisherUniversidad Nacional de Colombiaspa
dc.publisher.branchUniversidad Nacional de Colombia - Sede Bogotáspa
dc.publisher.departmentDepartamento de Ingeniería de Sistemas e Industrialspa
dc.publisher.facultyFacultad de Ingenieríaspa
dc.publisher.programBogotá - Ingeniería - Maestría en Ingeniería - Ingeniería de Sistemas y Computaciónspa
dc.relation.referencesAreeyapinan, Jennisa ; Kanongchaiyos, Pizzanu: Face morphing using critical point filters. En: JCSSE 2012 - 9th International Joint Conference on Computer Science and Software Engineering (2012), p. 283–288. ISBN 9781467319218spa
dc.relation.referencesAssa, Jackie ; Cohen-Or, Daniel: More of the same: Synthesizing a variety by structural layering. En: Computers and Graphics (Pergamon) 36 (2012), Nr. 4, p. 250–256. ISSN 00978493spa
dc.relation.referencesBau, David ; Zhu, Jun-Yan ; Strobelt, Hendrik ; Zhou, Bolei ; Tenenbaum, Joshua B. ; Freeman, William T. ; Torralba, Antonio: GAN Dissection: Visualizing and Understanding Generative Adversarial Networks. (2018)spa
dc.relation.referencesBergmann, Urs ; Jetchev, Nikolay ; Vollgraf, Roland: Learning Texture Manifolds with the Periodic Spatial GAN. (2017), mayspa
dc.relation.referencesBrock, Andrew ; Donahue, Jeff ; Simonyan, Karen: Large Scale GAN Training for High Fidelity Natural Image Synthesis. (2018), sep, p. 1–35spa
dc.relation.referencesCai, Lei ; Wang, Zhengyang ; Gao, Hongyang ; Shen, Dinggang ; Ji, Shuiwang: Deep Adversarial Learning for Multi-Modality Missing Data Completion. En: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining - KDD ’18. New York, New York, USA : ACM Press, jul 2018. – ISBN 9781450355520, p. 1158–1166spa
dc.relation.referencesChowdhury, S. M. A. K. ; Lubna, J. I.: Review on Deep Fake: A looming Technological Threat. En: 2020 11th International Conference on Computing, Communication and Networking Technologies (ICCCNT), 2020, p. 1–7spa
dc.relation.referencesDirvanauskas ; Maskeliunas ¯ ; Raudonis ; Damaˇsevicius ˇ ; Scherer: HEMIGEN: Human Embryo Image Generator Based on Generative Adversarial Networks. En: Sensors 19 (2019), aug, Nr. 16, p. 3578spa
dc.relation.referencesDong, Hao ; Zhang, Jingqing ; McIlwraith, Douglas ; Guo, Yike: I2T2I: Learning text to image synthesis with textual data augmentation. En: 2017 IEEE International Conference on Image Processing (ICIP) Vol. 2017-Septe, IEEE, sep 2017. – ISBN 9781–5090–2175–8, p. 2015–2019spa
dc.relation.referencesDong, Hao ; Zhang, Jingqing ; McIlwraith, Douglas ; Guo, Yike: I2T2I: Learning text to image synthesis with textual data augmentation. En: Proceedings - International Conference on Image Processing, ICIP 2017-Septe (2018), p. 2015–2019. – ISBN 9781509021758spa
dc.relation.referencesDu, Changde ; Du, Changying ; Huang, Lijie ; He, Huiguang: Reconstructing Perceived Images From Human Brain Activities With Bayesian Deep Multiview Learning. En: IEEE Transactions on Neural Networks and Learning Systems 30 (2018), Nr. 8, p. 2310–2323. – ISSN 21622388spa
dc.relation.referencesLi, Chuan ; Wand, Michael: Combining Markov Random Fields and Convolutional Neural Networks for Image Synthesis. (2016), janspa
dc.relation.referencesLi, Ming ; Ye, Chunyang ; Li, Wei: High-Resolution Network for Photorealistic Style Transfer. (2019), apr, Nr. 2001, p. 1–14spa
dc.relation.referencesLucic, Mario ; Kurach, Karol ; Michalski, Marcin ; Bousquet, Olivier ; Gelly, Sylvain: Are Gans created equal? A large-scale study. En: Advances in Neural Information Processing Systems 2018-December (2018), Nr. Nips, p. 700–709. – ISSN 10495258spa
dc.relation.referencesLucic, Mario ; Tschannen, Michael ; Ritter, Marvin ; Zhai, Xiaohua ; Bachem, Olivier ; Gelly, Sylvain: High-Fidelity Image Generation With Fewer Labels. (2019), marspa
dc.relation.referencesLy, Nam T. ; Nguyen, Cuong T. ; Nakagawa, Masaki: Training an End-to-End Model for Offline Handwritten Japanese Text Recognition by Generated Synthetic Patterns. En: 2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR) Vol. 2018-Augus, IEEE, aug 2018. – ISBN 978–1–5386–5875–8, p. 74–79spa
dc.relation.referencesMalik, Abid ; Lu, Micheal ; Wang, Nathenial ; Lin, Yeiwei ; Yoo, Shinjae: Detailed Performance Analysis of Distributed Tensorflow on a GPU Cluster using Deep Learning Algorithms. En: 2018 New York Scientific Data Summit (NYSDS), IEEE, aug 2018. – ISBN 978–1–5386–7933–3, p. 1–8spa
dc.relation.referencesMansimov, Elman ; Parisotto, Emilio ; Ba, Jimmy L. ; Salakhutdinov, Ruslan: Generating Images from Captions with Attention. (2015), p. 1–12 62 Bibliografíaspa
dc.relation.referencesMao, Xudong ; Li, Qing: Unpaired Multi-Domain Image Generation via Regularized Conditional GANs. En: Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence. California : International Joint Conferences on Artificial Intelligence Organization, jul 2018. – ISBN 9780999241127, p. 2553–2559spa
dc.relation.referencesMiyauchi, Yutaro ; Sugano, Yusuke ; Matsushita, Yasuyuki: Shape-Conditioned Image Generation by Learning Latent Appearance Representation from Unpaired Data. En: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol. 11366 LNCS. Springer Verlag, 2019. – ISBN 9783030208752, p. 438–453spa
dc.relation.referencesMizginov, V. A. ; Danilov, S. Y.: Synthetic thermal background and object texture generation using geometric information and GaN. En: International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives Vol. 42, International Society for Photogrammetry and Remote Sensing, 2019. – ISSN 16821750, p. 149–154spa
dc.relation.referencesPark, Taesung ; Liu, Ming-Yu ; Wang, Ting-chun ; Zhu, Jun-yan: Semantic Image Synthesis with Spatially-Adaptive Normalization. (2019), marspa
dc.relation.referencesRobertson, Scott ; Bertling, Thomas: How To draw. Desing Studio, 2013spa
dc.relation.referencesSangkloy, Patsorn ; Burnell, Nathan ; Ham, Cusuh ; Hays, James: The sketchy database. En: ACM Transactions on Graphics 35 (2016), Nr. 4, p. 1–12. – ISSN 07300301spa
dc.relation.referencesSannidhan, M. S. ; Ananth Prabhu, G. ; Robbins, David E. ; Shasky, Charles: Evaluating the performance of face sketch generation using generative adversarial networks. En: Pattern Recognition Letters 128 (2019), p. 452–458. – ISSN 01678655spa
dc.relation.referencesSiarohin, Aliaksandr ; Sangineto, Enver ; Lathuiliere, Stephane ; Sebe, Nicu: Deformable GANs for Pose-based Human Image Generation. En: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (2017), dec, p. 3408–3416. – ISBN 9781538664209spa
dc.relation.referencesSimonyan, Karen ; Zisserman, Andrew. Very Deep Convolutional Networks for LargeScale Image Recognition. 2015spa
dc.relation.referencesTalukdar, J. ; Gupta, S. ; Rajpura, P. S. ; Hegde, R. S.: Transfer Learning for Object Detection using State-of-the-Art Deep Neural Networks. En: 2018 5th International Conference on Signal Processing and Integrated Networks (SPIN), IEEE, feb 2018. – ISBN 978–1–5386–3045–7, p. 78–83 Bibliograf´ıa 63spa
dc.relation.referencesTan, Wei R. ; Chan, Chee S. ; Aguirre, Hernan ; Tanaka, Kiyoshi: Improved ArtGAN for Conditional Synthesis of Natural Image and Artwork. (2017), aug, p. 1–16spa
dc.relation.referencesTheagarajan, Rajkumar ; Bhanu, Bir: DeepHesc 2.0: Deep generative multi adversarial networks for improving the classification of hESC. En: PLoS ONE 14 (2019), mar, Nr. 3. – ISSN 19326203spa
dc.relation.referencesUlyanov, Dmitry ; Vedaldi, Andrea ; Lempitsky, Victor: Instance Normalization: The Missing Ingredient for Fast Stylization. (2016), jul, Nr. 2016spa
dc.relation.referencesWang, Haohan ; Raj, Bhiksha: On the Origin of Deep Learning. (2017), p. 1–72spa
dc.relation.referencesWu, Huikai ; Zheng, Shuai ; Zhang, Junge ; Huang, Kaiqi: GP-GAN: Towards Realistic High-Resolution Image Blending. (2017), mar. ISBN 9781450368896spa
dc.relation.referencesXie, Ning ; Hachiya, Hirotaka ; Sugiyama, Masashi: Artist agent: A reinforcement learning approach to automatic stroke generation in oriental ink painting. En: IEICE Transactions on Information and Systems E96-D (2013), Nr. 5, p. 1134–1144. – ISBN 9781450312851spa
dc.relation.referencesXie, Weidi ; Noble, J. A. ; Zisserman, Andrew: Microscopy cell counting and detection with fully convolutional regression networks. En: Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization 6 (2018), may, Nr. 3, p. 283–292. – ISSN 21681171spa
dc.relation.referencesYang, Wei ; Ouyang, Wanli ; Wang, Xiaolong ; Ren, Jimmy ; Li, Hongsheng ;spa
dc.relation.referencesWang, Xiaogang: 3D Human Pose Estimation in the Wild by Adversarial Learning. En: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (2018), p. 5255–5264. – ISBN 9781538664209spa
dc.relation.referencesYu, Bingqing ; Clark, James J.: WAYLA - Generating Images from Eye Movements. En: 2018 15th Conference on Computer and Robot Vision (CRV), IEEE, may 2018. – ISBN 978–1–5386–6481–0, p. 118–125spa
dc.relation.referencesZha, Xuewei ; Shi, Fei ; Ma, Yuhui ; Zhu, Weifang ; Chen, Xinjian: Generation of retinal OCT images with diseases based on cGAN. En: Angelini, Elsa D. (Ed.) ; Landman, Bennett A. (Ed.): Medical Imaging 2019: Image Processing, SPIE, mar 2019. – ISBN 9781510625457, p. 74spa
dc.relation.referencesZhang, Zhifei ; Song, Yang ; Qi, Hairong: Decoupled Learning for Conditional Adversarial Networks. En: 2018 IEEE Winter Conference on Applications of Computer Vision (WACV) Vol. 2018-Janua, IEEE, mar 2018. – ISBN 978–1–5386–4886–5, p. 700–708 64 Bibliografíaspa
dc.relation.referencesZhao, Bo ; Meng, Lili ; Yin, Weidong ; Sigal, Leonid: Image Generation from Layout. (2018), novspa
dc.relation.referencesZhu, Jun-Yan ; Park, Taesung ; Isola, Phillip ; Efros, Alexei A.: Unpaired Imageto-Image Translation Using Cycle-Consistent Adversarial Networks. En: 2017 IEEE International Conference on Computer Vision (ICCV) Vol. 2017-Octob, IEEE, oct 2017. – ISBN 978–1–5386–1032–9, p. 2242–2251spa
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.rights.licenseAtribución-CompartirIgual 4.0 Internacionalspa
dc.rights.urihttp://creativecommons.org/licenses/by-sa/4.0/spa
dc.subject.ddc000 - Ciencias de la computación, información y obras generales::003 - Sistemasspa
dc.subject.proposalBocetos Miniaturaspa
dc.subject.proposalRedes Neuronalesspa
dc.subject.proposalGANspa
dc.subject.proposalGeneración de Artespa
dc.subject.proposalThumbnail Arteng
dc.subject.proposalNeural Networkeng
dc.subject.proposalArt Generationeng
dc.titleDesarrollo de un prototipo de software de generación de bocetos miniatura con redes neuronalesspa
dc.title.translatedDevelopment of a software prototype forThumbnail generation with neural networks
dc.typeTrabajo de grado - Maestríaspa
dc.type.coarhttp://purl.org/coar/resource_type/c_bdccspa
dc.type.coarversionhttp://purl.org/coar/version/c_ab4af688f83e57aaspa
dc.type.contentTextspa
dc.type.driverinfo:eu-repo/semantics/masterThesisspa
dc.type.redcolhttp://purl.org/redcol/resource_type/TMspa
dc.type.versioninfo:eu-repo/semantics/acceptedVersionspa
oaire.accessrightshttp://purl.org/coar/access_right/c_abf2spa

Archivos

Bloque original

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
1032425183.2020.pdf
Tamaño:
27.9 MB
Formato:
Adobe Portable Document Format
Descripción:
Tesis de Maestría en Ingeniería de Sistemas y Computación

Bloque de licencias

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
license.txt
Tamaño:
3.87 KB
Formato:
Item-specific license agreed upon to submission
Descripción: