English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

It is well established in neuroscience that color vision plays an essentialpart in the human visual perception system. Meanwhile, many novel designs forcomputer vision inspired by human vision have achieved success in a wide rangeof tasks and applications. Nonetheless, how color differences affect machinevision has not been well explored. Our work tries to bridge this gap betweenthe human color vision aspect of visual recognition and that of the machine. Toachieve this, we curate two datasets: CIFAR10-F and CIFAR100-F, which are basedon the foreground colors of the popular CIFAR datasets. Together with CIFAR10-Band CIFAR100-B, the existing counterpart datasets with information on thebackground colors of CIFAR test sets, we assign each image based on its colorcontrast level per its foreground and background color labels and use this as aproxy to study how color contrast affects machine vision. We first conduct aproof-of-concept study, showing the effect of color difference and validate ourdatasets. Furthermore, on a broader level, an important characteristic of humanvision is its robustness against ambient changes; therefore, drawinginspirations from ophthalmology and the robustness literature, we analogizecontrast sensitivity from the human visual aspect to machine vision andcomplement the current robustness study using corrupted images with ourCIFAR-CoCo datasets. In summary, motivated by neuroscience and equipped withthe datasets we curate, we devise a new framework in two dimensions to performextensive analyses on the effect of color contrast and corrupted images: (1)model architecture, (2) model size, to measure the perception ability ofmachine vision beyond total accuracy. We also explore how task complexity anddata augmentation play a role in this setup. Our results call attention to newevaluation approaches for human-like machine perception.

On Human Visual Contrast Sensitivity and Machine Vision Robustness: A Comparative Study