Acquisition position
Tapio Wirkkala · Art glass · Uppsala Auktionskammare
Acquisition position
Potential market value
Evidence strength
Market anomaly
Risk profile
Investigation rationale
Decision guardrails
Visual laboratory
Object record
H ca 16 cm.
Identity resolution
Ranked from independent evidence. Confirmation or rejection becomes controlled training data.
Visual 0% · Name 86% · Marks 25% · Material 40%
33% combined confidenceVisual 0% · Name 42% · Marks 25% · Material 40%
26% combined confidenceVisual 0% · Name 41% · Marks 25% · Material 40%
25% combined confidenceVisual 0% · Name 41% · Marks 25% · Material 40%
25% combined confidenceVisual 0% · Name 40% · Marks 25% · Material 40%
25% combined confidenceVisual 0% · Name 39% · Marks 25% · Material 40%
25% combined confidenceVisual 0% · Name 37% · Marks 25% · Material 40%
25% combined confidenceVisual 0% · Name 36% · Marks 25% · Material 40%
24% combined confidenceVisual 0% · Name 35% · Marks 25% · Material 40%
24% combined confidenceVisual 0% · Name 35% · Marks 25% · Material 40%
24% combined confidenceVisual 0% · Name 35% · Marks 25% · Material 40%
24% combined confidenceVisual 0% · Name 34% · Marks 25% · Material 40%
24% combined confidenceClosed deal feedback
Enter the verified hammer price after the auction closes. Only verified outcomes become calibration data.
Learning guardrail
Global research evidence
Irving Penn American, 1917–2009
0% similarityCheck object rights; API metadata is openArt Institute of ChicagoLudwig Karl Hilberseimer American, born Germany, 1885–1967
0% similarityCheck object rights; API metadata is openArt Institute of ChicagoBerenice Abbott American, 1898–1991
0% similarityCheck object rights; API metadata is openArt Institute of ChicagoIvan Albright (American, 1897–1983)
0% similarityCheck object rights; API metadata is openArt Institute of ChicagoLen Prince (American, born 1953) Jessie Mann (American, born 1982)
0% similarityCheck object rights; API metadata is openArt Institute of ChicagoJan Kip (Dutch, c. 1653-1722) after Leonard Knyff (Dutch, 1650-1721) published by David Mortier (French, 1673-1728)
0% similarityCheck object rights; API metadata is openArt Institute of ChicagoJohn D. Graham American, born Russia (now Ukraine), 1886–1961
0% similarityCheck object rights; API metadata is openArt Institute of ChicagoFrancisco José de Goya y Lucientes Spanish, 1746-1828
0% similarityCheck object rights; API metadata is openArt Institute of ChicagoGeorges Rouault French, 1871-1958
0% similarityCheck object rights; API metadata is openArt Institute of ChicagoLisa Larsen American, 1925–1959
0% similarityCheck object rights; API metadata is openArt Institute of ChicagoRobert Goodnough American, 1917-2010
0% similarityCheck object rights; API metadata is openArt Institute of ChicagoFanny Lovejoy (1856/57) Weston, Vermont, United States
0% similarityCheck object rights; API metadata is openComparable logic
| Comparable ID | Match | Price | Reasons |
|---|---|---|---|
| 10655 | 68% | 500.0 SEK | identity 100%, material 45%, size 62%, title 82%, factory marks 45% |
| 10520 | 67% | 3766.0 SEK | identity 100%, material 45%, size 62%, title 72%, factory marks 45% |
| 10504 | 67% | 3766.0 SEK | identity 100%, material 45%, size 62%, title 72%, factory marks 45% |
| 16616 | 67% | 1200.0 SEK | identity 100%, material 45%, size 62%, title 70%, factory marks 45% |
| 19513 | 67% | 350.0 SEK | identity 100%, material 45%, size 62%, title 67%, factory marks 45% |
| 14828 | 66% | 900.0 SEK | identity 100%, material 45%, size 62%, title 60%, factory marks 45% |
| 11771 | 66% | 1133.0 SEK | identity 100%, material 45%, size 62%, title 61%, factory marks 45% |
| 8054 | 51% | 300.0 SEK | identity 45%, material 45%, size 62%, title 100%, factory marks 45% |
Automatic signature module
The local engine detects and isolates candidate regions, searches the trusted mark library and runs conservative signature forensics. Optional AI interprets unreadable regions and creates research queries without changing the forensic verdict.
The module runs during normal analysis when a local image is available.
Marks laboratory
110 independent methods generate cleanup variants, stroke and symbol descriptors, library matches, trust controls and conflict warnings.
Signature forensics
This does not declare a signature genuine. It measures fit against a multi-source reference set, confusing signers and natural variation.
Run mark matching first, then compare the questioned signature against a sufficiently broad trusted reference set.