Graph-based methods for the automatic annotation and retrieval of art prints

Abstract

The analysis of images taken from cultural heritage artifacts is an emerging area of research in the field of information retrieval. Current methodologies are focused on the analysis of digital images of paintings for the tasks of forgery detection and style recognition. In this paper, we introduce a graph-based method for the automatic annotation and retrieval of digital images of art prints. Such method can help art historians analyze printed art works using an annotated database of digital images of art prints. The main challenge lies in the fact that art prints generally have limited visual information. The results show that our approach produces better results in a weakly annotated database of art prints in terms of annotation and retrieval performance compared to state-of-the-art approaches based on bag of visual words.H.3.3Information storage and retrievalInformation search and retrieval|Retrieval models I.4.8Image Processing and Computer VisionScene Analysis|Object Recognition G.3Probability and StatisticsProbabilistic Algorithms General TermsAlgorithms, Measurement, Experimentation Keywords Content-based image retrieval and annotation, Art image annotation and retrieval, Graph-based Learning methods

Museum as a visual lab? - culture-technology networks in CBIR projects

“…Research studies instigated and continued by Carneiro on his own (Carneiro 2011;Carneiro et al 2012;Carneiro 2013) and in collaboration with Chen (Chen and Carneiro 2015) proposed two parallel research avenues, one being the automation of producing global annotations to previously unseen test images, the other focused on enabling retrieval of an un-annotated image from the database, given specifi c visual classes. Technically, this method follows a graph-based learning algorithm, relying on the assumption that visually similar images are likely to share the same annotations.…”

“…The PRINTART database introduced in 2010 (Carneiro 2011) comprised 307 annotated images with one multiclass problem and 21 binary problems, all images being collected from the Artstor digital image library (The Artstor database) and annotated manually by art historians. The selection of art prints was constrained by date (15 th -17 th century) and subject matter (religious theme).…”

“…Example of the manual annotation of an art print in the PRINTART database. Theme (multiclass problem) and the binary classes found in the image are listed (Carneiro 2011) Images in the PRINTART database are represented with the bag of visual words model (BOV), where each visual word is formed using a collection of local descriptors extracted by means of the scale invariant feature transform (SIFT).…”

An Early Framework for Determining Artistic Influence

“…In Computer Vision, there has been extensive research on the object-recognition in images, similarity between images. Also there has been united research on automated classification of paintings [1,2,3,9,8]. However, there is very little research done on measuring and determining influence between artists ,e.g.…”

Labelset anchored subspace ensemble (LASE) for multi-label annotation

“…To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. In multimedia information retrieval, annotating [3] [18] given texts, web data, music, images or videos with concepts or keywords plays a key role in browsing, ranking, indexing, search and navigation. Recently, the rapid rise of data volumes makes the efficiency and effectiveness of manual annotation unacceptable, and triggers a growing need for automatical annotation techniques, which are mainly developed from machine learning methods.…”