Reproduce “From the Drosophila Visual Connectome to General-Purpose Computer Vision” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zongyu Li, Akito Yamauchi, Huaizhi Liu, Vishwanatha Rao, Jia Guo, Frontotemporal Lobar Degeneration Neuroimaging Initiative, Alzheimer's Disease Neuroimaging Initiative
FlyVision is a connectome-informed visual architecture combining ON/OFF pathways, recurrent processing, and interactions among neuron-like populations. It scales across task-specific outputs and is evaluated on image classification and scan-level MRI prediction. On a 3,175-image chest test split, pretrained FlyVision Base achieved 92.60% accuracy versus 91.56% for ResNet18. These results support transfer across evaluated settings, not universal generality or biological causation.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.