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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Reconstructing perceived natural images from fMRI signals is one of the mostengaging topics of neural decoding research. Prior studies had success inreconstructing either the low-level image features or the semantic/high-levelaspects, but rarely both. In this study, we utilized an Instance-ConditionedGAN (IC-GAN) model to reconstruct images from fMRI patterns with both accuratesemantic attributes and preserved low-level details. The IC-GAN model takes asinput a 119-dim noise vector and a 2048-dim instance feature vector extractedfrom a target image via a self-supervised learning model (SwAV ResNet-50);these instance features act as a conditioning for IC-GAN image generation,while the noise vector introduces variability between samples. We trained ridgeregression models to predict instance features, noise vectors, and densevectors (the output of the first dense layer of the IC-GAN generator) ofstimuli from corresponding fMRI patterns. Then, we used the IC-GAN generator toreconstruct novel test images based on these fMRI-predicted variables. Thegenerated images presented state-of-the-art results in terms of capturing thesemantic attributes of the original test images while remaining relativelyfaithful to low-level image details. Finally, we use the learned regressionmodel and the IC-GAN generator to systematically explore and visualize thesemantic features that maximally drive each of several regions-of-interest inthe human brain.

Reconstruction of Perceived Images from fMRI Patterns and Semantic Brain Exploration using Instance-Conditioned GANs