107:
20:
302:
approach used by Art
Recognition. Notably, a crucial difference emerged in the datasets used to train the respective AI systems. While the Bradford group's AI was trained using 49 images, Art Recognition utilized a substantially larger dataset of over 100 images. This difference in the size and composition of the training datasets underscored the significant impact that data selection has on the outcomes of AI-driven art analysis.
322:: A painting attributed to Titian, housed at Kunsthaus Zürich, has been a topic of debate among art experts. The application of Art Recognition's technology offered a new perspective, utilizing AI to analyze the painting's stylistic elements in comparison with authenticated works of Titian. Following this debate,
331:
In each of these instances, Art
Recognition's involvement has provided additional perspectives through AI analysis while contributing to broader conversations about the role of technology in art authentication. These cases demonstrate the evolving nature of art verification, where traditional methods
255:
The technology developed by Art
Recognition has been recognized for its role in providing a technology-based art authentication solution, compared to traditional methods. This advancement is seen as significant in the field of art verification, offering a modern approach to a historically complex
93:
These technological advancements, particularly in the realm of high-resolution digital imagery, enable a more detailed examination of artworks. By analyzing brushstrokes, signature patterns, and other distinct characteristics, and comparing them with known works by the same artist, digital tools
301:
initially attributed the painting to
Raphael, employing an AI face recognition software, while the AI developed at Art Recognition returned a negative result. As the face recognition method proved inadequate for art authentication, the Bradford group developed a new technology more akin to the
266:
Despite the advancements in AI-driven art authentication, the field continues to face unique challenges, particularly regarding the acceptance of such technologies. Experts in the field stress the necessity of using AI as a complementary tool alongside traditional methods, rather than as a
326:
has announced plans to initiate a comprehensive project aimed at resolving the authenticity questions surrounding the painting. This project is set to involve collaboration with scientists and technology companies, leveraging a multidisciplinary approach to authenticate the
312:, a painting attributed to Lucian Freud became a subject of dispute. Art Recognition's AI analysis played a pivotal role in examining the painting's authenticity, contributing to the broader discussion about the challenges in verifying modern artworks.
152:
Upon the preparation of datasets, a segment of the image set is used for training the AI algorithm, while the remaining images are set aside for testing. This phase aims to ensure the algorithm's proficiency in distinguishing authentic artworks from
81:
of authenticating paintings, a process traditionally reliant on expert judgment, historical research, and scientific analysis. Recognizing the limitations of existing methods, the co-founders were motivated by technological advancements in
539:
263:, where experts deliberated over the capabilities and limitations of AI in identifying art forgeries. Such discussions highlight the evolving landscape of art authentication in the age of digital technology.
290:, has also been examined using Art Recognition's AI, which has assessed the painting as non-authentic. This analysis contributed to ongoing scholarly discussions regarding the work's authenticity.
130:. The AI algorithm analyzes various visual characteristics, such as brushstrokes, color palette, texture, and composition, to identify patterns and similarities with known authentic artworks.
164:
After the testing phase, the AI algorithm is applied to analyze new images, including submissions from clients. Additionally, the algorithm is designed to identify artworks generated by
145:
to ensure the inclusion of genuine artworks by specific artists. This approach aims to avoid including artworks that may have been partially completed by apprentices or contain mixed
94:
offer a new dimension in authentication. Popovici and Hoppe-Oehl aimed to develop an advanced algorithm that could further assist experts by identifying stylistic elements and
332:
are being supplemented, and sometimes challenged, by new technological approaches. However, they also underline the ongoing debates about the acceptance of AI in the field of
77:
Art
Recognition was established in 2019 by Dr. Carina Popovici and Christiane Hoppe-Oehl. The foundation of the company was driven by the long-standing challenge in the
259:
The use of AI in art authentication, as pioneered by Art
Recognition, has become a topic of professional discourse. Notably, this subject was the focus of a debate on
110:
The AI Report includes an AI-determined authenticity probability, analytical heatmaps, brushstroke visualizations, and outlines the methodology and historical context.
275:
Art
Recognition's AI algorithm has been applied to several high-profile and controversial artworks, sparking significant interest and debate in the art world.
803:
590:
564:
220:
Art
Recognition's AI algorithm has received attention from various media outlets and industry events. The company was featured on the front page of
353:
808:
676:"Two A.I. Models Produced Different Results When Authenticating a Raphael Painting. Here's Why That Doesn't Undermine the Tool's Potential"
623:
507:
481:
457:
838:
818:
205:
752:
656:
675:
753:"A Zurich Museum Found Out It May Have Acquired a Fake Titian. So Why Did It Buy Another Painting That Looks Just Like It?"
828:
193:
165:
719:
65:
technology. The company's operations extend globally, with a primary aim to increase transparency and security in the
224:
for its involvement in the authentication case of the Flaget
Madonna, believed to have been partly painted by
106:
172:, enhancing its reliability in differentiating between authentic and artificially generated fake art pieces.
260:
169:
279:
700:
433:
823:
244:
221:
197:
141:, and training. In the initial phase, datasets are compiled, and data selection is supervised by
50:
46:
236:
158:
833:
298:
189:
168:, mimicking the style of renowned artists. This capability equips the algorithm to withstand
208:
Impact
Acceleration Award. Furthermore, Art Recognition has established a relationship with
813:
8:
294:
127:
123:
87:
323:
319:
19:
390:
181:
565:"AI Companies Are Authenticating Old Master Paintings, But the Art World is Skeptical"
286:: The National Gallery's "Samson and Delilah", traditionally attributed to the artist
157:. Post-training, the algorithm undergoes evaluation with the test data, assessing its
727:
631:
598:
515:
361:
287:
38:
283:
138:
115:
409:
212:, a Swiss innovation agency, to expand its research and development initiatives.
134:
119:
83:
62:
410:"2312.14998 - Synthetic images aid the recognition of human-made art forgeries"
309:
201:
781:
209:
196:) funding program. In addition, the company has formed a partnership with the
797:
731:
635:
602:
519:
365:
98:
unique to individual artists, thus aiding in the art authentication process.
305:
235:
showcased how the algorithm can be used to detect art forgeries with high
458:"Is This a Real Raphael Painting? AI Says Yes, But Humans Aren't So Sure"
333:
185:
142:
54:
42:
146:
66:
540:"L'intelligence artificielle peut-elle détecter les faux dans l'art?"
240:
78:
58:
591:"Was famed Samson and Delilah really painted by Rubens? No, says AI"
35:
508:"Art Recognition: Carina Popovici legt Kunstfälschern das Handwerk"
434:"Innosuisse Discover 2021 - Recognising art forgeries from a photo"
395:
188:
has resulted in the acquisition of a research grant from Eurostars,
31:
336:, especially in the authentication of works by renowned artists.
225:
154:
95:
624:"Painting 'undoubtedly' by Raphael to go on display in Bradford"
385:
Schaerf, Ludovica; Popovici, Carina; Postma, Eric (2023-07-10),
315:
788:
701:"AI meets Old Masters in the fight to authenticate paintings"
657:"A question of attribution: just how useful can AI tools be?"
232:
482:"Die Idee - Mit einem Algorithmus Kunstfälschungen erkennen"
267:
stand-alone or definitive solution for authenticating art.
354:"New tools are making it easier to authenticate paintings"
192:
the Eureka's flagship small and medium-sized enterprises (
16:
Technology company headquartered in Adliswil, Switzerland
384:
239:. Additionally, the company's work was featured in a
175:
795:
206:Science and Technology Facilities Council (STFC)
133:The company's technology undergoes a process of
293:De Brecy Tondo Madonna. A research team from
101:
387:Art Authentication with Vision Transformers
231:A broadcast by the Swiss public television
250:
394:
215:
161:and efficacy in authenticating artworks.
114:Art Recognition employs a combination of
30:is a technology company headquartered in
673:
105:
18:
804:Companies based in the canton of Zürich
720:"The Case of the Disputed Lucian Freud"
621:
588:
796:
717:
654:
505:
270:
57:, Art Recognition integrates advanced
750:
180:Art Recognition's collaboration with
45:. Specializing in the application of
809:Swiss companies established in 2019
486:Schweizer Radio und Fernsehen (SRF)
13:
839:2019 establishments in Switzerland
698:
308:Painting Controversy: Featured in
204:, which has been supported by the
14:
850:
819:Artificial intelligence companies
773:
176:Academic partnerships and grants
90:algorithms in the field of art.
744:
711:
692:
674:Popovici, Carina (2023-10-24).
667:
648:
615:
582:
557:
532:
499:
474:
450:
426:
402:
378:
346:
1:
655:Harris, Gareth (2023-10-24).
622:Khomami, Nadia (2023-07-14).
589:Alberge, Dalya (2021-09-26).
339:
751:Dafoe, Taylor (2023-03-28).
506:Müller, André (2020-01-19).
243:discussing the use of AI in
7:
10:
855:
829:21st-century introductions
718:Knight, Sam (2022-09-19).
284:National Gallery in London
120:computer vision algorithms
102:Technology and methodology
72:
49:(AI) for the purposes of
514:(in Swiss High German).
128:authenticity of artworks
546:(in French). 2021-12-19
488:(in German). 2020-10-23
261:Radio Télévision Suisse
251:Debates and discussions
222:The Wall Street Journal
198:University of Liverpool
47:artificial intelligence
216:Recognition and impact
111:
24:
299:Nottingham University
190:Eureka (organisation)
109:
53:and the detection of
22:
789:Art magazine website
512:Neue Zürcher Zeitung
414:www.emergentmind.com
124:deep neural networks
705:The Financial TImes
295:Bradford University
271:Controversial cases
170:adversarial attacks
139:dataset preparation
88:pattern recognition
280:Samson and Delilah
245:art authentication
182:Tilburg University
112:
51:art authentication
25:
661:The Art Newspaper
39:metropolitan area
846:
785:
784:
782:Official website
767:
766:
764:
763:
748:
742:
741:
739:
738:
715:
709:
708:
699:Harris, Gareth.
696:
690:
689:
687:
686:
671:
665:
664:
652:
646:
645:
643:
642:
619:
613:
612:
610:
609:
586:
580:
579:
577:
576:
561:
555:
554:
552:
551:
536:
530:
529:
527:
526:
503:
497:
496:
494:
493:
478:
472:
471:
469:
468:
454:
448:
447:
445:
444:
430:
424:
423:
421:
420:
406:
400:
399:
398:
382:
376:
375:
373:
372:
350:
324:Kunsthaus Zürich
320:Kunsthaus Zürich
116:machine learning
854:
853:
849:
848:
847:
845:
844:
843:
794:
793:
780:
779:
776:
771:
770:
761:
759:
749:
745:
736:
734:
716:
712:
697:
693:
684:
682:
672:
668:
653:
649:
640:
638:
620:
616:
607:
605:
587:
583:
574:
572:
563:
562:
558:
549:
547:
538:
537:
533:
524:
522:
504:
500:
491:
489:
480:
479:
475:
466:
464:
456:
455:
451:
442:
440:
432:
431:
427:
418:
416:
408:
407:
403:
383:
379:
370:
368:
352:
351:
347:
342:
273:
253:
218:
178:
135:data collection
104:
84:digital imaging
75:
63:computer vision
28:Art Recognition
17:
12:
11:
5:
852:
842:
841:
836:
831:
826:
824:Authentication
821:
816:
811:
806:
792:
791:
786:
775:
774:External links
772:
769:
768:
743:
724:The New Yorker
710:
691:
666:
647:
614:
581:
556:
531:
498:
473:
449:
425:
401:
377:
344:
343:
341:
338:
329:
328:
313:
310:The New Yorker
303:
291:
272:
269:
252:
249:
217:
214:
202:United Kingdom
177:
174:
143:art historians
126:to assess the
103:
100:
74:
71:
15:
9:
6:
4:
3:
2:
851:
840:
837:
835:
832:
830:
827:
825:
822:
820:
817:
815:
812:
810:
807:
805:
802:
801:
799:
790:
787:
783:
778:
777:
758:
754:
747:
733:
729:
725:
721:
714:
706:
702:
695:
681:
677:
670:
662:
658:
651:
637:
633:
629:
625:
618:
604:
600:
596:
592:
585:
570:
566:
560:
545:
541:
535:
521:
517:
513:
509:
502:
487:
483:
477:
463:
459:
453:
439:
435:
429:
415:
411:
405:
397:
392:
388:
381:
367:
363:
359:
358:The Economist
355:
349:
345:
337:
335:
325:
321:
317:
314:
311:
307:
304:
300:
296:
292:
289:
285:
281:
278:
277:
276:
268:
264:
262:
257:
248:
246:
242:
238:
234:
229:
227:
223:
213:
211:
207:
203:
199:
195:
191:
187:
183:
173:
171:
167:
166:generative AI
162:
160:
156:
150:
148:
144:
140:
136:
131:
129:
125:
121:
117:
108:
99:
97:
91:
89:
85:
80:
70:
68:
64:
60:
56:
55:art forgeries
52:
48:
44:
40:
37:
34:, within the
33:
29:
21:
834:Data science
760:. Retrieved
756:
746:
735:. Retrieved
723:
713:
704:
694:
683:. Retrieved
679:
669:
660:
650:
639:. Retrieved
628:The Guardian
627:
617:
606:. Retrieved
595:The Observer
594:
584:
573:. Retrieved
571:. 2023-03-01
568:
559:
548:. Retrieved
543:
534:
523:. Retrieved
511:
501:
490:. Retrieved
485:
476:
465:. Retrieved
461:
452:
441:. Retrieved
437:
428:
417:. Retrieved
413:
404:
386:
380:
369:. Retrieved
357:
348:
330:
306:Lucian Freud
274:
265:
258:
254:
230:
219:
179:
163:
151:
132:
118:techniques,
113:
92:
76:
27:
26:
814:Art forgery
757:Artnet News
680:Artnet News
334:art history
186:Netherlands
43:Switzerland
798:Categories
762:2024-02-10
737:2024-02-10
685:2024-02-10
641:2024-02-10
608:2024-02-10
575:2024-02-10
550:2023-06-23
525:2024-02-10
492:2024-02-10
467:2024-02-10
443:2024-02-10
438:InnoSuisse
419:2024-02-10
396:2307.03039
371:2024-02-10
340:References
210:Innosuisse
147:authorship
67:art market
59:algorithms
732:0028-792X
636:0261-3077
603:0029-7712
520:0376-6829
366:0013-0613
256:process.
241:TEDx talk
155:forgeries
79:art world
569:Observer
327:artwork.
237:accuracy
159:accuracy
96:patterns
32:Adliswil
282:at the
226:Raphael
200:in the
184:in The
73:History
730:
634:
601:
544:rts.ch
518:
364:
316:Titian
288:Rubens
122:, and
36:Zurich
391:arXiv
728:ISSN
632:ISSN
599:ISSN
516:ISSN
362:ISSN
297:and
86:and
61:and
23:Logo
462:WSJ
318:at
233:SRF
194:SME
800::
755:.
726:.
722:.
703:.
678:.
659:.
630:.
626:.
597:.
593:.
567:.
542:.
510:.
484:.
460:.
436:.
412:.
389:,
360:.
356:.
247:.
228:.
149:.
137:,
69:.
41:,
765:.
740:.
707:.
688:.
663:.
644:.
611:.
578:.
553:.
528:.
495:.
470:.
446:.
422:.
393::
374:.
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.