FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366978032> ?p ?o ?g. }

• W4366978032 endingPage "244" @default.
• W4366978032 startingPage "228" @default.
• W4366978032 abstract "The contrast sensitivity function (CSF) is a fundamental signature of the visual system that has been measured extensively in several species. It is defined by the visibility threshold for sinusoidal gratings at all spatial frequencies. Here, we investigated the CSF in deep neural networks using the same 2AFC contrast detection paradigm as in human psychophysics. We examined 240 networks pretrained on several tasks. To obtain their corresponding CSFs, we trained a linear classifier on top of the extracted features from frozen pretrained networks. The linear classifier is exclusively trained on a contrast discrimination task with natural images. It has to find which of the two input images has higher contrast. The network’s CSF is measured by detecting which one of two images contains a sinusoidal grating of varying orientation and spatial frequency. Our results demonstrate characteristics of the human CSF are manifested in deep networks both in the luminance channel (a band-limited inverted U-shaped function) and in the chromatic channels (two low-pass functions of similar properties). The exact shape of the networks’ CSF appears to be task-dependent. The human CSF is better captured by networks trained on low-level visual tasks such as image-denoising or autoencoding. However, human-like CSF also emerges in mid- and high-level tasks such as edge detection and object recognition. Our analysis shows that human-like CSF appears in all architectures but at different depths of processing, some at early layers, while others in intermediate and final layers. Overall, these results suggest that (i) deep networks model the human CSF faithfully, making them suitable candidates for applications of image quality and compression, (ii) efficient/purposeful processing of the natural world drives the CSF shape, and (iii) visual representation from all levels of visual hierarchy contribute to the tuning curve of the CSF, in turn implying a function which we intuitively think of as modulated by low-level visual features may arise as a consequence of pooling from a larger set of neurons at all levels of the visual system." @default.
• W4366978032 created "2023-04-27" @default.
• W4366978032 creator A5004240998 @default.
• W4366978032 creator A5018435072 @default.
• W4366978032 creator A5084884328 @default.
• W4366978032 date "2023-07-01" @default.
• W4366978032 modified "2023-10-07" @default.
• W4366978032 title "Contrast sensitivity function in deep networks" @default.
• W4366978032 cites W1964640535 @default.
• W4366978032 cites W1971709186 @default.
• W4366978032 cites W1983976362 @default.
• W4366978032 cites W1988013646 @default.
• W4366978032 cites W1999908130 @default.
• W4366978032 cites W2003752843 @default.
• W4366978032 cites W2004242458 @default.
• W4366978032 cites W2006301684 @default.
• W4366978032 cites W2007205332 @default.
• W4366978032 cites W2018106756 @default.
• W4366978032 cites W2030493486 @default.
• W4366978032 cites W2035989480 @default.
• W4366978032 cites W2039960782 @default.
• W4366978032 cites W2040036684 @default.
• W4366978032 cites W2042925217 @default.
• W4366978032 cites W2046131422 @default.
• W4366978032 cites W2047872994 @default.
• W4366978032 cites W2053895730 @default.
• W4366978032 cites W2058616551 @default.
• W4366978032 cites W2060258698 @default.
• W4366978032 cites W2063445096 @default.
• W4366978032 cites W2066230840 @default.
• W4366978032 cites W2067045150 @default.
• W4366978032 cites W2067608582 @default.
• W4366978032 cites W2081881941 @default.
• W4366978032 cites W2096124620 @default.
• W4366978032 cites W2133665775 @default.
• W4366978032 cites W2136325353 @default.
• W4366978032 cites W2145889472 @default.
• W4366978032 cites W2146577751 @default.
• W4366978032 cites W2153782322 @default.
• W4366978032 cites W2166206801 @default.
• W4366978032 cites W2166278041 @default.
• W4366978032 cites W2212384750 @default.
• W4366978032 cites W2395611524 @default.
• W4366978032 cites W2412480261 @default.
• W4366978032 cites W2537084945 @default.
• W4366978032 cites W2774788508 @default.
• W4366978032 cites W2898929289 @default.
• W4366978032 cites W3006472786 @default.
• W4366978032 cites W3019981565 @default.
• W4366978032 cites W3030216326 @default.
• W4366978032 cites W3083092310 @default.
• W4366978032 cites W3122510310 @default.
• W4366978032 cites W3134644686 @default.
• W4366978032 cites W3159228069 @default.
• W4366978032 cites W3198661929 @default.
• W4366978032 cites W4220652501 @default.
• W4366978032 cites W4232132600 @default.
• W4366978032 cites W4236443542 @default.
• W4366978032 cites W4247010311 @default.
• W4366978032 cites W4280540175 @default.
• W4366978032 cites W4310572212 @default.
• W4366978032 cites W4311282550 @default.
• W4366978032 cites W4321466059 @default.
• W4366978032 cites W4361772798 @default.
• W4366978032 cites W76624669 @default.
• W4366978032 doi "https://doi.org/10.1016/j.neunet.2023.04.032" @default.
• W4366978032 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37156217" @default.
• W4366978032 hasPublicationYear "2023" @default.
• W4366978032 type Work @default.
• W4366978032 citedByCount "3" @default.
• W4366978032 countsByYear W43669780322023 @default.
• W4366978032 crossrefType "journal-article" @default.
• W4366978032 hasAuthorship W4366978032A5004240998 @default.
• W4366978032 hasAuthorship W4366978032A5018435072 @default.
• W4366978032 hasAuthorship W4366978032A5084884328 @default.
• W4366978032 hasBestOaLocation W43669780322 @default.
• W4366978032 hasConcept C100921725 @default.
• W4366978032 hasConcept C115961682 @default.
• W4366978032 hasConcept C120665830 @default.
• W4366978032 hasConcept C121332964 @default.
• W4366978032 hasConcept C127162648 @default.
• W4366978032 hasConcept C15123163 @default.
• W4366978032 hasConcept C153180895 @default.
• W4366978032 hasConcept C154945302 @default.
• W4366978032 hasConcept C160086991 @default.
• W4366978032 hasConcept C16345878 @default.
• W4366978032 hasConcept C169760540 @default.
• W4366978032 hasConcept C196956537 @default.
• W4366978032 hasConcept C2524010 @default.
• W4366978032 hasConcept C26760741 @default.
• W4366978032 hasConcept C2776502983 @default.
• W4366978032 hasConcept C2984842247 @default.
• W4366978032 hasConcept C31258907 @default.
• W4366978032 hasConcept C31972630 @default.
• W4366978032 hasConcept C33923547 @default.
• W4366978032 hasConcept C41008148 @default.
• W4366978032 hasConcept C50644808 @default.
• W4366978032 hasConcept C73313986 @default.

• next