Stork Data Errors 2
The figures below supplement the information at Stork Error Summary and Stork Data Errors.
• David G. Stork, “Did Jan van Eyck build the first ‘photocopier’ in 1432?” SPIE Electronic Imaging Color Imaging IX: Processing, Hardcopy, and Applications, R. Eschbach and G. G. Marcu (eds.) pp. 50-56, 2004.
Stork claims in this paper that his data shows the relative shifts of the images are orthogonal, in a ratio of 1:2. However, he has misregistered both images by several mm when he overlapped them. This is outside of experimental uncertainty. The actual data from properly registered images shows them to be non-orthogonal and in a ratio of ~1:1.
• Thomas Ketelsen, Olaf Simon, Ina Reiche, Silke Merchel, and David G. Stork, “Evidence for mechanical (not optical) copying and enlarging in Jan van Eyck’s Portrait of Niccolò Albergati,” Proceedings of the Optical Society of America Annual Meeting 2004.
and
• David G. Stork, “Optics and realism in Renaissance art,” Scientific American, 291(6):76-84, December 2004.
Left) Stork’s fit of a copy made with a proportional divider (Green) to van Eyck’s drawing (outlined in Red). Discrepancies over the entire area of the image show that Stork’s fit is 2.6% too large.
Right) This figure shows a significantly improved fit over the entire image when the copy is reduced by 2.6% with respect to the enlargement Stork used. This difference is well outside of experimental error. Below are details of the fits in the region of the ear.
Stork claims in this paper that his fit of an enlargement of a copy made with a proportional divider is excellent, except in the area of the ear. However, outside of experimental error, his fit was made with the drawing 2.6% too large. The center inset (Magenta) shows that his fit is consistently incorrect over the region shown here. The overall fit to all parts of the image, including the ear, is better when the enlargement is reduced by 2.6%.
• David G. Stork, “Optics and the Old Masters Revisited,” Optics and Photonics News, 15(3), pp. 30-37, March 2004.
Left) Outside of experimental error, Stork’s fit of the perspective lines to a photograph of a modern reproduction of a chandelier, and the horizon line he extracts from his fit, are significantly different than actual data from this photograph. The actual data shows the camera was tilted by ~3o when the photograph was taken, and that the arms deviate from the perfect symmetry Stork incorrectly reports from his data.
Right) Here I show to the correct relative scale the chandelier Stork used for his comparison and the one in van Eyck’s painting. Stork’s selected a chandelier that is a modern reproduction, much smaller and less complex than van Eyck’s, and with only four arms instead of six. All of these factors make it a deceptive choice for Stork to use for comparison with a painting of a large, complex, six arm, hand-made 15th century chandelier.
• Antonio Criminisi and David G. Stork, “Did the great masters use optical projections while painting? Perspective comparison of paintings and photographs of Renaissance chandeliers,” in J. Kittler, M. Petrou and M. S. Nixon (eds.), Proceedings of the 17th International Conference on Pattern Recognition, Volume IV, pp. 645-648, 2004.
and
• David G. Stork, “Optics and realism in Renaissance art,” Scientific American, 291(6):76-84, December 2004.
Partially hidden from view is a figure from our 2003 paper (which can be downloaded from my website), showing in Green the features we had identified as optics-based, and in Blue the features we had identified as non-optical. As I’ve shown in Green and Blue on one of the figures in Stork and Criminisi’s paper, they found the same result as we had the previous year. However, they selectively omitted without discussion these (Green) optical features, basing their incorrect conclusion that optics had not been used on only the (Blue) features we had previously identified as non-optical. Also, they claim credit for the the analysis technique (deconstructing the 3D chandelier into its six individual 2D arms) we had introduced the previous year. In fact, their results reproduce, rather than contradict, ours.
• David G. Stork, “Were optical projections used in early Renaissance painting? A geometric vision analysis of Jan van Eyck’s ‘Arnolfini portrait’ and Robert Campin’s ‘MĂ©rode Altarpiece’,” SPIE Electronic Imaging, Vision Geometry XII, L. J. Latecki, D. M. Mount and A. Y. Wu (eds), pp. 23-30, 2004.
Stork shows in his paper that every slat “is in fact perfectly straight.” This data is false, outside any experimental error. The figure shows in Yellow and Blue where 5–8o kinks are in the slats that slope up to the left, and in Green and Cyan where similar kinds are in the slats that slope down to the left. The large inset shows an expanded view of one set of kinks.
• David G. Stork, “Asymmetry in ‘Lotto carpets’ and implications for Hockney’s optical projection theory,” SPIE Electronic Imaging, San Jose, January, 2005.
and
• Christopher W. Tyler and David G. Stork, “Did Lorenzo Lotto use optical projections when painting Husband and wife?,” Proceedings of the Optical Society of America Annual Meeting 2004.
Left) Of the only 5 carpets shown in an overview essay that Stork incorrectly refers to as “Mack’s study,” two are similar to Lotto’s. Of those two, Stork selected the one that is clearly most distorted from 500 years of wear, water damage, etc.
Right) The smaller green rectangle shows the region of this carpet that Stork selected to show in his paper. Had he shown the entire carpet, it would be obvious to the reader that it was too distorted to use as data. On the right of this figure I’ve simply “morphed” the outer border of the carpet back to the roughly rectangular shape that it had when new. This action removed the apparent asymmetry of the octagonal pattern that Stork incorrectly used for his conclusions. His conclusion is based on selection of data that is quite obviously unrepresentative and incorrect.