Special Sessions


Computer Graphics in the Study of Master Paintings and Drawings: Three-Dimensional Virtual Worlds Behind the Two-Dimensional Masterpieces

Friday, 17 December | 4:15 PM - 6:00 PM | Auditorium

Presented in English, translated simultaneously to Korean / 영어로 발표 됨 (한국어 동시 통역)

Computer graphics models of artists' studios allow scholars to address a number of vexing problems in the history and interpretation of art, particularly those related to old masters' working methods. Recent studies have used this modeling technique to analyze how some of the world's most famous paintings were created:

• A model of Jan van Eyck's Portrait of Arnolfini and his wife exposed perspective inaccuracies that are difficult to discern by the unaided eye. The model also confirmed that the focal length of the convex mirror in the painting was much shorter than a putative projection mirror for this work. Both results led to rejection of the claim that this work was executed by tracing optical projections.

• The lighting direction estimated in a model of Jan Vermeer's Girl With a Pearl Earring agreed closely with directions estimated by five other sources within the painting, objectively revealing this artist's mastery in rendering the effects of light and supporting the claim that this work was executed with a live model, not from the artist's imagination.

• A model of Georges de la Tour's Christ in the Carpenter's Studio confirmed that the light in the tableau was at the candle, rather than "in place of the other figure", and this, in turn, led to a rejection of the claim that this painting was executed using optical projections.

• A model of Parmigianino's Self Portrait in a Convex Mirror revealed that the warped image is consistent with the artist faithfully recording the image of a rectilinear room distorted by the mirror, rather than inventing a fictive space. The model also reveals that the work may be hung too high in its gallery home.

This talk, profusely illustrated with art works and movies of computer graphics models of artists' studios, explains the steps by which computer experts and art scholars build such computer graphics models, the types of assumptions that are brought to bear, and the strengths and limitations of the overall methodology. The talk concludes with a number of unsolved problems in the history of art that seem amenable to these computer graphics methods.

David G. Stork
Ricoh Innovations

David G. Stork, Chief Scientist, Ricoh Innovations, has held academic appointments in physics, mathematics, electrical engineering, statistics, computer science, neuroscience, psychology, and art and art history at Wellesley and Swarthmore Colleges and Clark, Boston, and Stanford Universities. He is a Fellow of the International Association for Pattern Recognition and Chair of its Technical Committee on Computer Vision in Cultural Heritage Applications. He has published six books or proceedings and has another in production, including Seeing the Light: Optics in Nature, Photography, Color, Vision and Holography (Wiley), the leading textbook on optics in the arts; the first three volumes in the disciplines of computer vision and computer graphics in the study of art; Pattern Classification (2nd ed., Wiley), the world's all-time best-selling textbook in the field, translated into three languages and used in courses in over 250 universities worldwide; and HAL's Legacy: 2001's Computer as Dream and Reality (MIT), the source of his PBS television documentary "2001: HAL's Legacy". He holds 40 US patents and has published numerous technical papers on human and machine learning and perception of patterns, physiological optics, image understanding, concurrency theory, theoretical mechanics, optics, image processing and computer graphics. His past and confirmed schedule includes over 220 scholarly presentations on computer methods in the study of fine art in 18 countries.