Transfer Learning for Illustration Classification (Unknown language)

How to get this document?

Download
Commercial Copyright fee: €14.50 Basic fee: €4.00 Total price: €18.50
Academic Copyright fee: €4.50 Basic fee: €2.00 Total price: €6.50

The field of image classification has shown an outstanding success thanks to the development of deep learning techniques. Despite the great performance obtained, most of the work has focused on natural images ignoring other domains like artistic depictions. In this paper, we use transfer learning techniques to propose a new classification network with better performance in illustration images. Starting from the deep convolutional network VGG19, pre-trained with natural images, we propose two novel models which learn object representations in the new domain. Our optimized network will learn new low-level features of the images (colours, edges, textures) while keeping the knowledge of the objects and shapes that it already learned from the ImageNet dataset. Thus, requiring much less data for the training. We propose a novel dataset of illustration images labelled by content where our optimized architecture achieves 86.61% of top-1 and 97.21% of top-5 precision. We additionally demonstrate that our model is still able to recognize objects in photographs.

Table of contents conference proceedings

The tables of contents are generated automatically and are based on the data records of the individual contributions available in the index of the TIB portal. The display of the Tables of Contents may therefore be incomplete.

1
An Improved Parallel Technique for Neighbour Search on CUDA
Perea, Juan J. / Cordero, Juan M. | 2017
11
Dissipation Potentials for Yarn-Level Cloth
Sánchez-Banderas, Rosa M. / Otaduy, Miguel A. | 2017
19
Physically Based Skeleton Tracking
López-Gandía, Axel / Susín, Antonio | 2017
23
Extending Industrial Digital Twins with Optical Object Tracking
Tammaro, Antonio / Segura, Álvaro / Moreno, Aitor / Sánchez, Jairo R. | 2017
27
Unsupervised Framework for People Counting Using a Stereo-based Camera
Negrillo, José / Feito, Francisco R. / Segura, Rafael J. / Ogayar, Carlos Javier / Fuertes, José Manuel / Lucena, Manuel | 2017
31
Google Tango Outdoors. Augmented Reality for Underground Infrastructures
Soria, Gregorio / Ortega, Lidia / Feito, Francisco R. | 2017
41
Direct Volume Rendering of Stack-Based Terrains
Graciano, Alejandro / Rueda, Antonio J. / Feito, Francisco R. | 2017
51
Downsampling and Storage of Pre-Computed Gradients for Volume Rendering
Díaz-García, Jesús / Brunet, Pere / Navazo, Isabel / Vázquez, Pere-Pau | 2017
61
3D GIS Based on WebGL for the Management of Underground Utilities
Jurado, Juan Manuel / Ortega, Lidia / Feito, Francisco R. | 2017
65
Fireman Rescue: A Serious Game for Fire Fighting Training
Ríos, Alejandro / Bonet, Carles / Morales, J. L. / Alavedra, Axel / París, Alejandro / Guillén, Marc | 2017
69
A Curvature-based Method for Identifying the Contact Zone Between Bone Fragments: First Steps
Jiménez-Pérez, J. Roberto / Paulano-Godino, Félix / Jiménez-Delgado, Juan J. | 2017
73
An Interactive Tool for Modeling Ancient Masonry Buildings
Fita, Josep Lluis / Besuievsky, Gonzalo / Patow, Gustavo | 2017
77
Transfer Learning for Illustration Classification
Lagunas, Manuel / Garces, Elena | 2017
87
Fast Stippling based on Weighted Centroidal Voronoi Diagrams
Gómez, Eila / Méndez, Elías / Arroyo, Germán / Martín, Domingo | 2017
97
Improved Intuitive Appearance Editing based on Soft PCA
Malpica, Sandra / Barrio, Miguel / Gutierrez, Diego / Serrano, Ana / Masia, Belen | 2017
107
Transient Photon Beams
Marco, Julio / Jarosz, Wojciech / Gutierrez, Diego / Jarabo, Adrian | 2017
113
Procedural Semantic Cities
Rogla, Otger / Pelechano, Nuria / Patow, Gustavo | 2017
121
Tree Variations
Argudo, Oscar / Andújar, Carlos / Chica, Antoni | 2017
131
Procedural Generation of Natural Environments with Restrictions
Gasch, Cristina / Chover, Miguel / Remolar, Inmaculada | 2017
Feedback