Séminaire ETIS : Rick Johnson & David Picard, Son Vu, Inbar Fijalkow

Titre du séminaire et orateurs

  • "Signal Processing in Computational Art History", Rick Johnson, Cornell University, USA.
  • "Photographic Paper Texture Classification using Model Deviation of Local Visual Descriptors", présenté par David Picard, Son Vu et Inbar Fijalkow, ETIS.

Date et lieu

Mardi 17 juin 2014, 15h.
ENSEA, amphi A.


Signal Processing in Computational Art History

Ce séminaire sera aussi diffusé sur Ximinds.

A key component of the scholarly analysis of fine art, which recently has expanded significantly under the label of technical art history, utilizes the extraction of features from revealing images of the art object. Within the past seven years painstaking manual methods of feature extraction from images of an art work’s support materials, in particular canvas and paper, have been enhanced with the application of signal processing in projects spearheaded by the speaker. When combined with big data handling capabilities, creating an approach designated here as computational art history, significant advances have been achieved. This talk describes three such emerging applications of digital signal processing to art historical issues of dating and attribution:

  • thread counting for weave matching of Old Master paintings on canvas from x-radiographs, which has proven helpful in studies of the paintings of van Gogh, Vermeer, Velazquez, Bouts, and a Poussin
  • texture similarity assessment for metadata (manufacturer, surface finish, brand, and date of manufacture) classification of historic photographic papers, in particular silver gelatin and inkjet papers, from raking light photomicrographs
  • chain line pattern matching for mold-mate identification of laid papers from low-energy radiographs, initially for the prints of Rembrandt and Durer

tl_files/site-etis/docs-actu/images/RickJohnson.jpgC. Richard Johnson, Jr. is the Geoffrey S. M. Hedrick Senior Professor of Engineering and a Stephen H. Weiss Presidential Fellow at Cornell University. He received a PhD in Electrical En- gineering from Stanford University, along with the first PhD minor in Art History granted by Stanford, in 1977. At the start of 2007, after 30 years of research on adaptive feedback systems theory and blind equalization in communication receivers, Professor Johnson accepted a 5-year appointment as an Adjunct Research Fellow of the Van Gogh Museum (Amsterdam, the Netherlands) to facilitate the interaction of art historians and conservation specialists with algorithm-building signal processors. In 2013, Professor Johnson was appointed a Scientific Researcher of the Rijksmuseum (Amsterdam, the Netherlands) and Computational Art History Advisor to the RKD - Netherlands Institute for Art History (the Hague, the Netherlands). Professor Johnson founded the Thread Count Automation Project (TCAP) in collaboration with the Van Gogh Museum in 2007, initiated the Historic Photographic Paper Classification (HPPC) Challenge in cooperation with the Museum of Modern Art in 2010, and launched the Chain Line Pattern (CLiP) Matching Project with the Morgan Library & Museum and the Rijksmuseum in 2012.

Photographic Paper Texture Classification using Model Deviation of Local Visual Descriptors

This paper investigates the classification of photographic paper textures using visual descriptors. Such classification is called fine grain due to the very low inter-class variability. We propose a novel image representation for photographic paper texture categorization, relying on the incorporation of a powerful local descriptor into an efficient higher-order model deviation where texture is represented by computing statistics on the occurrences of specific local visual patterns. We perform an evaluation on two different challenging datasets of photographic paper textures and show such advanced methods indeed outperforms existing descriptors alone.

This contribution will be presented at ICIP 2014.