BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:The Valencia-Bristol low-level vision MindSet - Jesús Malo
DTSTART:20250711T133000Z
DTEND:20250711T140000Z
UID:TALK233896@talks.cam.ac.uk
CONTACT:Yancheng Cai
DESCRIPTION:Recent claims about the superiority of deep-nets to model huma
 n vision are based on the reproduction of either (a) neural recordings fro
 m different brain layers or (b) high-level behaviors included in Brain-Sco
 re. However\, high correlations in such tasks do not guarantee the functio
 nal similarity of the underlying mechanisms in models and humans. In parti
 cular\, appart from exceptions (like your work \;-)\, not many people is l
 ooking at the bottleneck of artificial systems as characterized by low-lev
 el visual psychophysics. In this talk we present stimuli for 20 different 
 experiments that highlight basic color/texture/motion perception facts\, a
 nd how the trends of the artificial responses can be used to assess the si
 milarities with human vision.\n\nRecommended reading:\n* Li\, Gomez-Villa\
 , Bertalmío & Malo\nContrast sensitivity functions in autoencoders\nJourn
 al of Vision (May 2022)\, Vol.22\, 8. doi:https://doi.org/10.1167/jov.22.6
 .8\n\n* Malo\, Vila-Tomas\, Hernandez-Camara\, Li\, Laparra\nA Turing Test
  for Artificial Nets devoted to model Human Vision (june 2022)\nhttp://isp
 .uv.es/docs/talk_AI_Bristol_Malo_et_al_2022.pdf\n\n* Malo & Bowers\nThe Lo
 w-Level vision MindSet.\nTalk at the Seminars on Generalisation in Mind an
 d Machine. (sept. 2024)\nhttp://isp.uv.es/docs/Low_Level_MindSet_3.pptx\n\
 nZoom link: https://cam-ac-uk.zoom.us/j/81459588387?pwd=Uyq625q1QSKJ15ZvvS
 nzAbV44NTf1w.1\n
LOCATION:SS03 - William Gates Building
END:VEVENT
END:VCALENDAR
