rgeirhos/generalisation-humans-DNNs

Data, code & materials from the paper "Generalisation in humans and deep neural networks" (NeurIPS 2018)

44
/ 100
Emerging

This project provides the data and materials for researchers studying how well deep neural networks generalize compared to human vision. It includes a curated dataset of images, code for manipulating these images with various distortions, and the raw accuracy data from both human participants and several neural networks. Cognitive scientists, vision scientists, and AI researchers can use this to compare new models against established benchmarks of human perception.

No commits in the last 6 months.

Use this if you want to analyze or benchmark the generalization performance of your image recognition models against human perception data from controlled psychology experiments.

Not ideal if you are looking for a plug-and-play solution for training new deep learning models without needing to delve into psychological experimental data.

cognitive-science computer-vision-benchmarking psychophysics human-perception AI-generalization
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 19 / 25

How are scores calculated?

Stars

95

Forks

22

Language

R

License

Last pushed

Aug 14, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/rgeirhos/generalisation-humans-DNNs"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.