MonkeySee

MonkeySee: Space-time-resolved reconstructions of natural images from macaque multi-unit activity

Read the full paper

In this project, we reconstructed naturalistic images directly from macaque brain signals using a convolutional neural network (CNN)-based decoder. We explored how different brain regions, like V1, V4, and IT, contribute to image decoding. This research helps bridge the gap between low-level and high-level brain signals, enabling high-precision reconstructions and revealing new insights into how the brain’s representations are mapped into visual pixels.

Key Highlights:

  • CNN-based decoding for brain signals
  • Space-time-resolved decoding technique
  • Investigation of different brain regions’ contributions
  • High-precision natural image reconstructions

Publication: