Pandora: Description of a Painting Database for Art Movement Recognition with Baselines and Perspectives

To facilitate computer analysis of visual art, in the form of paintings, we introduce Pandora (Paintings Dataset for Recognizing the Art movement) database, a collection of digitized paintings labelled with respect to the artistic movement. Noting that the set of databases available as benchmarks for evaluation is highly reduced and most existing ones are limited in variability and number of images, we propose a novel large scale dataset of digital paintings. The database consists of more than 7700 images from 12 art movements. Each genre is illustrated by a number of images varying from 250 to nearly 1000. We investigate how local and global features and classification systems are able to recognize the art movement. Our experimental results suggest that accurate recognition is achievable by a combination of various categories.To facilitate computer analysis of visual art, in the form of paintings, we introduce Pandora (Paintings Dataset for Recognizing the Art movement) database, a collection of digitized paintings labelled with respect to the artistic movement. Noting that the set of databases available as benchmarks for evaluation is highly reduced and most existing ones are limited in variability and number of images, we propose a novel large scale dataset of digital paintings. The database consists of more than 7700 images from 12 art movements. Each genre is illustrated by a number of images varying from 250 to nearly 1000. We investigate how local and global features and classification systems are able to recognize the art movement. Our experimental results suggest that accurate recognition is achievable by a combination of various categories.

READ FULL TEXT VIEW PDF
POST COMMENT

Comments

There are no comments yet.

Authors

page 5

03/27/2017

A Study on the Extraction and Analysis of a Large Set of Eye Movement Features during Reading

This work presents a study on the extraction and analysis of a set of 10...
03/31/2016

Exemplar-AMMs: Recognizing Crowd Movements from Pedestrian Trajectories

In this paper, we present a novel method to recognize the types of crowd...
11/06/2011

Face Recognition Using Discrete Cosine Transform for Global and Local Features

Face Recognition using Discrete Cosine Transform (DCT) for Local and Glo...
11/26/2019

A Two-stream End-to-End Deep Learning Network for Recognizing Atypical Visual Attention in Autism Spectrum Disorder

Eye movements have been widely investigated to study the atypical visual...
12/06/2020

Art Style Classification with Self-Trained Ensemble of AutoEncoding Transformations

The artistic style of a painting is a rich descriptor that reveals both ...
08/24/2021

Pen Spinning Hand Movement Analysis Using MediaPipe Hands

We challenged to get data about hand movement in pen spinning using Medi...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.

1 Introduction

The remarkable expansion of the digital data during the last period favored a much easier access to works of art for the general public. Great efforts were put lately into creating automatic image processing solutions that facilitate a better understanding of art [Cornelis et al.(2011)Cornelis, Dooms, Cornelis, Leen, and Schelkens]. These solutions may aim at obtaining high-quality and high-fidelity digital versions of paintings [Martinez et al.(2002)Martinez, Cupitt, Saunders, and Pillay] or may address various aspects such as: image diagnostics, virtual restoration, color rejuvenation etc. as discussed in the review of Stork et al. [Stork(2009)]. Another more appropriate to the ultimate goal of computers is the context recognition. One of the broadest possible implementation of context recognition is the automatic art movement identification.

According to Artyfactory [Art(Retrieved February 2016)], art movements are “collective titles that are given to artworks which share the same artistic ideals, style, technical approach or timeframe”. While some works are clearly set into a single art movement, others are in the transition period, as painters loved to experiment new ideas, leading to creation of a new movement. Also while the actual characteristics place a work in some art movement, its author, for personal reasons, refused to be categorized in such a way, giving birth to disputes.

In this paper, we look into the problem of computational categorization of digitized paintings into artistic genres (or art movements). In contrast to other directions of image classification, such as scene or object recognition, where large databases and evaluation protocols do exist, such an aspect is less emphasized for digitized paintings. Typically, the evaluation of a new method is carried on a small database with few paintings belonging to few genres. Given the latest advances of machine learning, two aspects should be noted: (1) deep networks with the many parameters easily overfit on small databases and (2) to have progress, we need larger databases.

In this paper we start by reviewing painting collections introduced in prior art and we follow by describing the proposed database. Next, to form a baseline, we continue by reporting the performance of various popular image descriptors and machine learning systems on the introduced database. The paper ends with discussions and conclusions.

2 Related work

In the last period multiple solutions issued automatic analysis of visual art and especially paintings using computer vision techniques. However, most of the research is based on medium–to–small databases. A summary of such methods is presented in table

1. One may easily note the size of the databases (and implicitly the number of art movements investigated) increased with time, while the reported performance decreased until it stabilized in the range of 50-70% for correct art movement recognition. Some of the most representative databases used for art movement identification are:

Concluding, many of the databases previously used, are small and contain non-standard evaluation protocols allowing overfitting. Thus, a larger scale database with fixed evaluation protocol should be beneficial for further development on the topic.

Method
Move-
ments
Db.
size
RR. Test ratio
Gunsel et al. [Gunsel et al.(2005)Gunsel, Sariel, and Icoglu] 3 107 91.66% 53.5%
Zujovic et al. [Zujovic et al.(2009)Zujovic, Gandy, Friedman, Pardo, and Pappas] 5 353 68.3% 10% CV
Siddiquie et al. [Siddiquie et al.(2009)Siddiquie, Vitaladevuni, and Davis] 6 498 82.4% 20% CV
Shamir et al. [Shamir et al.(2010)Shamir, Macura, Orlov, Eckley, and Goldberg] 3 517 91% 29.8%
Arora&ElGammal[Arora et al.(2012)Arora, , and Elgammal] 7 490 65.4% 20% CV
Khan et al. [Khan et al.(2014)Khan, Beigpour, van de Weijer, and Felsberg] 13 2338 62.2% 46.53% CV
Condorovici et al.[Condorovici et al.(2015)Condorovici, Florea, and Vertan] 8 4119 72.24% 10% CV
Agarwal et al. [Agarwal et al.(2015)Agarwal, Karnick, Pant, and Patel] 10 3000 62.37% 10% CV
Proposed 12 7724 54.7% 25% CV

Table 1: Art movement recognition solutions with the size of used databases. The database size refers only to the database used for art movement recognition, as in some cases larger databases have been implied for other purposes. The value for recognition rate (RR) is the one reported by the respective work while the “test ratio” is the percentage used for testing from the overall database (CV-stands for cross validation). We kindly ask the reader to retrieve all the details from the respective work.

3 Pandora database

Our main contribution is the creation of a new and extensive dataset of art images111The up-to-date database with pre-computed features data reported here is available at http://imag.pub.ro/pandora/pandora_download.html . While we follow the Paintings-91 database [Khan et al.(2014)Khan, Beigpour, van de Weijer, and Felsberg], our dataset is significantly larger, it was build around art movements and not painters and we tried to span wider time periods from antiquity to current periods. The later property should help the automatic study of style evolution, of thematic evolution and cross-time relationship identifications.

The Pandora (Paintings Dataset for Recognizing the Art movement) dataset consists of 7724 images from 12 movements: old Greek pottery, iconoclasm, high renaissance, baroque, rococo, romanticism, impressionism, realism, cubism, fauvism, abstract-expressionism and surrealism. The precise database structure is shown in table 2 and some examples representative for the art movements are in figure 1. We kindly ask the reader to note some of difficulties in distinguishing between genres: the main difference between Abstract and Fauvism is the less natural order in the structure of the Abstract works, while the Fauvism tends “to use color to express joy“. Baroque has a darker tone with respect to Romanticism while the later depicts "exotism or extraordinary things" . The difference between Realism and Surrealism is that the later illustrate “irrational juxtaposition of images” [Art(Retrieved February 2016)] (e.g. such as wings attached to the girl). Yet thinking in computer terms, to detect the irrational of joy in an image is extremely hard. Thus we consider that to achieve such goals, one needs, first, an appropriate database of considerable size and variability.

Art movement No. of paintings Historical period
Old Greek pottery 350 Antiquity
Iconoclasm 665 Middle age
High renaissance 812 1490 - 1527
Baroque 960 1590 - 1725
Rococo 844 1650 - 1850
Romanticism 874 1770 - 1850
Impressionism 984 1860 - 1925
Realism 307 1848 - present
Cubism 920 1900 - present
Abstract-expressionism 340 1920 - present
Fauvism 426 1900 - 1950
Surrealism 242 1900 - present
Table 2: The structure of the Pandora database.
Old Greek Pottery Orthodox Iconoclasm High Renaissance Baroque
Rococo Romanticism Impressionism Realism
Cubism Fauvism Abstract Surrealism
Figure 1: The 12 art movements illustrated in the proposed database.

4 Art movement recognition performance

4.1 Training and testing

To separate the database training and testing parts, a 4-fold cross validation scheme was implemented. The division into 4 folds exists at the level of each art movement, thus each image being uniquely allocated into a fold. The same division was used for all further tests and it is part of the database.

4.2 Features and classifiers

As “there is no fixed rule that determines what constitutes an art movement” and ”the artists associated with one movement may adhere to strict guiding principles, whereas those who belong to another may have little in common” [Art(Retrieved February 2016)], there cannot be a single set of descriptors that are able to separate any two art movements.

Following the observations from prior works [Arora et al.(2012)Arora, , and Elgammal], [Khan et al.(2014)Khan, Beigpour, van de Weijer, and Felsberg], multiple categories of feature descriptors should be used. For instance, to differentiate between impressionism and previous styles, one of the main difference is the brush stroke, thus texture. Old Pottery and Orthodox Iconoclasm are older and use a limited color palette. Also, one needs to understand the content of the painting to distinguish between realism and surrealism (for instance); thus, global composition descriptor should be used.

To provide a baseline for further evaluation, we have tested various combinations of popular feature extractors and classification algorithms.

The texture feature extractors used are :

The color descriptors tested are:

Machine learning classification systems tested are:

  • Support Vector Machine. We have relied for its implementation on the Lib-SVM [Chang and Lin(2011)]

    . We used on the radial basis function c-SVM and followed, for each case, the optimization (i.e. exhaustive search in (cost,gamma) space) recommended by the LibSVM creators.

  • Random forest [Breiman(2001)]. We have used 100 trees and unlimited depth. At each node we randomly look for a split in dimensions where is the input feature dimension.

    Let us note that before the development of the deep networks the random forests and support vector machines have been found to be the most robust families of classifiers

    [Fernández-Delgado et al.(2014)Fernández-Delgado, Cernadas, Barro, and Amorim]. Also, for small and diverse databases SVM and RF out-compete deep networks.

  • k-Nearest neighbor

    (kNN). We have implemented 1-NN, 3-NN and 7-NN based on Euclidean distance. While we report the results in terms of correct recognition rate, the nearest neighbor results will give an indication about the retrieval performance as it may be translated in terms of precision–recall.

Furthermore we have tested several systems that were previously used for art movement recognition. Inspired from previous work [Arora et al.(2012)Arora, , and Elgammal], we have run the Bag of Words (BoW) over SIFT keypoint detector with a vocabulary of 500. We have also tested a combination of color description, texture analysis based on Gabor filters and scene composition based on Gestalt frameworks [Condorovici et al.(2015)Condorovici, Florea, and Vertan].

Additionally, while the database is small for such a purpose and thus not really suited for deep learning, to have an indication of baseline performance, we have trained and evaluated a version of Deep Convolutional Neural Network (CNN). Our implementation is based on the MatConvNet

[Vedaldi and Lenc(2015)] library and LeNet architecture [LeCun et al.(1998)LeCun, Bottou, Bengio, and Haffner].

4.3 Results

We report first the results achieved when various combinations of features and classifiers are used (to be followed in table 3). We also report, in tables 4, 5, the confusion matrices for the best combination in each category: pLPB+SVM, GIST+RF, CSD+SVM and respectively pLBP+CSD+SVM.

Feat. / Class. Random Forest SVM 1-NN 3-NN 7-NN
HOG 0.266 0.248 0.200 0.214 0.233
pHOG 0.342 0.364 0.262 0.266 0.267
colorHOG 0.268 0.277 0.213 0.221 0.236
LBP 0.386 0.395 0.303 0.298 0.320
pLBP 0.459 0.525 0.368 0.362 0.377
LIOP 0.344 0.362 0.246 0.252 0.260
EHD 0.319 0.287 0.270 0.267 0.286
GIST 0.379 0.337 0.297 0.280 0.282
DCN 0.298 0.264 0.192 0.201 0.215
CSD 0.435 0.489 0.337 0.3357 0.363
pLBP + DCN 0.488 0.521 0.278 0.282 0.297
pLBP + CSD 0.540 0.547 0.377 0.282 0.297
Table 3: Recognition rates when various combinations of features and classifiers are used on the Pandora database. We marked with bold the best performance.
Gre Ren Icon Roc Rom Bar Impr Cub Abs Fauv Real Sur
Gre 200 20 28 2 3 5 35 51 3 1 0 2
Ren 5 360 24 54 47 153 95 70 1 2 0 1
Icon 7 21 559 7 1 15 23 30 2 0 0 0
Roc 4 90 21 272 129 214 73 38 1 0 0 2
Rom 9 106 16 165 219 183 107 55 2 5 3 4
Bar 4 173 21 186 121 292 87 65 2 3 0 6
Impr 10 97 47 66 106 95 429 109 8 6 2 9
Cub 14 93 65 42 55 76 167 361 12 12 2 21
Abs 11 43 43 5 18 11 66 67 46 2 0 28
Fauv 7 53 20 15 35 43 118 89 6 18 1 21
Real 2 10 16 22 25 17 39 16 1 3 135 21
Surr 2 13 16 7 16 7 43 37 14 7 2 78
GIST + RF
Gre Ren Icon Roc Rom Bar Impr Cub Abs Fauv Real Sur
Gre 339 4 1 1 0 3 0 1 0 0 1 0
Ren 0 404 16 49 66 219 24 31 0 2 1 0
Icon 0 8 576 3 0 6 30 31 0 11 0 0
Roc 0 118 2 230 153 262 51 18 0 1 8 1
Rom 0 91 4 113 322 228 74 31 0 1 10 0
Bar 0 178 8 152 150 405 42 18 0 1 5 1
Impr 0 24 58 33 35 66 584 133 13 29 9 0
Cub 2 43 42 38 21 32 168 508 16 30 5 15
Abs 0 3 11 1 7 2 55 83 125 45 0 8
Fauv 0 9 15 8 8 11 104 85 34 145 3 4
Real 1 18 14 30 29 42 69 38 4 7 50 5
Surr 0 3 5 8 16 5 44 65 15 19 5 57
CSD + SVM
Table 4: Confusion Matrices for the best performers on each category. The art movement used acronyms are: old Greek pottery–Gre, iconoclasm–Ico, high renaissance–Ren, baroque–Bar, rococo–Roc, romanticism–Rom, impressionism–Impr, realism–Real, cubism–cub, fauvism–Fauv, abstract-expressionism–Abs and surrealism–Sur.
Gre Ren Icon Roc Rom Bar Impr Cub Abs Fauv Real Sur
Gre 271 4 2 1 5 1 11 27 5 18 2 3
Ren 4 444 1 40 60 166 36 58 1 0 1 1
Icon 42 1 598 1 0 3 3 10 3 4 0 0
Roco 0 48 3 324 136 232 63 36 0 0 1 1
Rom 3 94 4 139 339 182 80 27 1 3 0 2
Bar 1 126 3 165 136 441 47 37 0 2 0 2
Impr 9 29 10 49 92 43 618 112 6 14 1 1
Cub 63 29 23 17 29 28 128 570 8 17 0 8
Abs 20 11 21 7 22 17 56 91 59 23 2 11
Fauv 33 5 5 1 18 9 111 58 11 163 2 10
Real 15 11 4 7 17 17 33 23 13 18 141 8
Surr 18 13 10 4 23 19 17 40 13 13 4 68
pLBP + SVM
Gre Ren Icon Roc Rom Bar Impr Cub Abs Fauv Real Sur
Gre 307 0 0 0 0 0 2 40 0 0 1 0
Ren 0 473 2 49 63 159 24 41 0 1 0 0
Icon 0 1 616 1 1 2 10 31 1 2 0 0
Roc 0 83 1 301 163 214 59 23 0 0 0 0
Rom 0 77 0 104 372 184 74 63 0 0 0 0
Bar 0 132 0 169 148 437 50 24 0 0 0 0
Impr 0 17 7 31 43 36 696 143 0 10 1 0
Cub 2 21 13 11 21 17 119 706 1 9 0 0
Abs 0 0 4 1 6 1 53 200 62 13 0 0
Fauv 0 2 2 2 6 1 126 190 3 93 1 0
Real 0 3 0 9 13 11 56 79 0 4 131 1
Surr 0 1 3 4 7 2 33 159 4 6 0 23
pLBP + CSD + SVM
Table 5: Confusion Matrices for the best performers on each category. The art movement used acronyms are: old Greek pottery–Gre, iconoclasm–Ico, high renaissance–Ren, baroque–Bar, rococo–Roc, romanticism–Rom, impressionism–Impr, realism–Real, cubism–cub, fauvism–Fauv, abstract-expressionism–Abs and surrealism–Sur.

Secondly we report comparatively the best performance of aggregated systems in table 6. We note that for this particular database, the best performance is achieved by a standard combination of features (pyramidal LBP + Color Structure Descriptor) with a Support Vector Machine.

While one may find disappointing the performance of various established systems, this is perfectly explainable. For the Bag of Words there is too much variability between keypoints to find a common ground; instead of the baseline version tested here, one should opt for much larger vocabularies with accurate compression to keep memory requirements low. Regarding the performance of the DeepCNN, the reported value should be perceived as a lower boundary, as the database is too small for directly training nets with tens of thousands of variables, since no data augmentation was implemented and the images being resized at lost some of the defining characteristics.

System Performance
pLBP + CSD +SVM 0.547
BoW 0.352
Condorovici et al. [Condorovici et al.(2015)Condorovici, Florea, and Vertan] 0.379
Deep CNN 0.486
Table 6: Recognition rates when various systems are used.

5 Discussion and conclusions

The best achieved performance was by a combination of pyramidal LBP and Color Structure Descriptor. One may expect the addition of GIST to further increase the performance, but this does not happen, probably due to the curse of dimensionality (the features dimension reaching 800); in such a case a feature selection method should be used, but we consider it outside the scope of the current paper.

The next important observation is that different descriptors do a good job separating some currents and not so good on identifying others. For instance, the CSD separates excellently the Orthodox Iconoclasm which has a unique color palette (due to degradation in time and reduced colors available at creation), but it is not able to separate Fauvism from Impressionism as both use the same colors but distributed differently. The Surrealism is hard to separate by everything else except GIST as it is the only tested feature able to describe the scene composition. Yet the GIST is not able to distinguish the Fauvism from Impressionism as local texture makes the difference. In contrast, the pLBP confusion between Fauvism and Impressionism is much reduced.

Overall, the confusion between Abstract and Cubism is large. As Cubism is defined by the extraordinary apparition of straight lines, to address it, one should try to introduce features appropriate to describe rectilinear objects.

Concluding we propose a new painting database annotated with art movements labels and divided in 4 folds to prepare it for rigorous evaluation. The database is significantly larger than the ones previously used. We have tested a multitude of popular features and classifiers and we have identified the weak and strong points of each of them. We also suggest some directions for future research that we anticipate to be beneficial for progress in the field.

Acknowledgment

This work is supported by a grant of the Romanian National Authority for Scientific Research and Innovation, CNCS UEFISCDI, number PN-II-RU-TE-2014-4-0733

References

  • [Art(Retrieved February 2016)] What is an art movement ? http://www.artyfactory.com/art_appreciation/art_movements/art_movements.htm, Retrieved February 2016.
  • [Agarwal et al.(2015)Agarwal, Karnick, Pant, and Patel] S. Agarwal, H. Karnick, N. Pant, and U. Patel. Genre and style based painting classification. In Proc. of WACV, pages 588–594, 2015.
  • [Arora et al.(2012)Arora, , and Elgammal] R. S. Arora, , and Ahmed Elgammal. Towards automated classification of fine–art painting style: a comparative study. In Proc. of ICPR, pages 3541–3544, 2012.
  • [Baştan et al.(2009)Baştan, Çam, Güdükbay, and Özgür Ulusoy] Muhammet Baştan, Hayati Çam, Uǧur Güdükbay, and Özgür Ulusoy. BilVideo-7: An MPEG-7-Compatible Video Indexing and Retrieval System. IEEE MultiMedia, 17(3):62–73, 2009.
  • [Breiman(2001)] Leo Breiman. Random forests. Machine learning, 45(1):5–32, 2001.
  • [Chang and Lin(2011)] Chih-Chung Chang and Chih-Jen Lin. Libsvm: A library for support vector machines. ACM Trans. Intell. Syst. Technol., 2(3), May 2011.
  • [Condorovici et al.(2015)Condorovici, Florea, and Vertan] Razvan George Condorovici, Corneliu Florea, and Constantin Vertan. Automatically classifying paintings with perceptual inspired descriptors. J. Vis. Commun. Image. Represent., 26:222 – 230, 2015.
  • [Cornelis et al.(2011)Cornelis, Dooms, Cornelis, Leen, and Schelkens] B. Cornelis, A. Dooms, J. Cornelis, F. Leen, and P. Schelkens. Digital painting analysis, at the cross section of engineering, mathematics and culture. In Proc. of EUSIPCO, pages 1254–1259, 2011.
  • [Dalal and Triggs(2005)] N. Dalal and B. Triggs. Histograms of oriented gradients for human detection. In Proc. of CVPR, pages 886–893, 2005.
  • [Fernández-Delgado et al.(2014)Fernández-Delgado, Cernadas, Barro, and Amorim] Manuel Fernández-Delgado, Eva Cernadas, Senén Barro, and Dinani Amorim. Do we need hundreds of classifiers to solve real world classification problems? JMLR, 15(1):3133–3181, 2014.
  • [Gunsel et al.(2005)Gunsel, Sariel, and Icoglu] B. Gunsel, S. Sariel, and O. Icoglu. Content-based access to art paintings. In Proc. of ICIP, pages 558–561, 2005.
  • [Huang and Wang(2014)] Yin-Fu Huang and Chang-Tai Wang. Classification of painting genres based on feature selection. In Proc. of Multimedia and Ubiquitous Engineering, LNEE, volume 308, pages 159–164, 2014.
  • [Khan et al.(2014)Khan, Beigpour, van de Weijer, and Felsberg] Fahad Shahbaz Khan, Shida Beigpour, Joost van de Weijer, and Michael Felsberg. Painting-91: a large scale database for computational painting categorization. Mach. Vis. App., 25(6):1385–1397, 2014.
  • [Khan et al.(2013)Khan, van de Weijer, Shahbaz Khan, Muselet, Ducottet, and Barat] R. Khan, J. van de Weijer, F. Shahbaz Khan, D. Muselet, C. Ducottet, and C. Barat. Discriminative color descriptors. In Proc. of CVPR, pages 2866–2873, 2013.
  • [LeCun et al.(1998)LeCun, Bottou, Bengio, and Haffner] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324, 1998.
  • [Manjunath et al.(2001)Manjunath, Ohm, Vasudevan, and Yamada] B. S. Manjunath, J. R. Ohm, V. V. Vasudevan, and A. Yamada. Color and texture descriptors. IEEE Trans. Cir. and Sys. for Video Technol., 11(6):703–715, 2001.
  • [Martinez et al.(2002)Martinez, Cupitt, Saunders, and Pillay] K. Martinez, J. Cupitt, D. Saunders, and R. Pillay. Ten years of art imaging research. Proceedings of the IEEE, 90(1):28–41, 2002.
  • [Ojala et al.(2002)Ojala, Pietikäinen, and Mäenpää] Timo Ojala, Matti Pietikäinen, and Topi Mäenpää. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. IEEE Trans. Pattern Anal. Mach. Intell., 24(7):971–987, July 2002.
  • [Oliva and Torralba(2001)] Aude Oliva and Antonio Torralba. Modeling the shape of the scene: A holistic representation of the spatial envelope. IJCV, 42(3):145–175, 2001.
  • [Shamir et al.(2010)Shamir, Macura, Orlov, Eckley, and Goldberg] Lior Shamir, Tomasz Macura, Nikita Orlov, D. Mark Eckley, and Ilya G. Goldberg. Impressionism, expressionism, surrealism: Automated recognition of painters and schools of art. ACM Trans Appl Percept, 7(2):1–17, 2010.
  • [Siddiquie et al.(2009)Siddiquie, Vitaladevuni, and Davis] B. Siddiquie, S.N. Vitaladevuni, and L.S. Davis. Combining multiple kernels for efficient image classification. In Proc. of WACV, pages 1–8, 2009.
  • [Stork(2009)] D. Stork. Computer vision and computer graphics analysis of paintings and drawings: An introduction to the literature. In Proc. of CAIP, pages 9–24, 2009.
  • [Vedaldi and Fulkerson(2010)] Andrea Vedaldi and Brian Fulkerson. Vlfeat: An open and portable library of computer vision algorithms. In Proc. of ACM MM, pages 1469–1472, 2010.
  • [Vedaldi and Lenc(2015)] Andrea Vedaldi and Karel Lenc. Matconvnet: Convolutional neural networks for matlab. In Proc. of ACM MM, pages 689–692, 2015.
  • [Wang et al.(2011)Wang, Fan, and Wu] Zhenhua Wang, Bin Fan, and Fuchao Wu. Local intensity order pattern for feature description. In Proc. of ICCV, pages 603–610, 2011.
  • [Zujovic et al.(2009)Zujovic, Gandy, Friedman, Pardo, and Pappas] J. Zujovic, L. Gandy, S. Friedman, B. Pardo, and T.N. Pappas. Classifying paintings by artistic genre: An analysis of features & classifiers. In Proc. of IEEE MMSP, pages 1–5, 2009.