rgeirhos/texture-vs-shape

Pre-trained models, data, code & materials from the paper "ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness" (ICLR 2019 Oral)

by Robert Geirhos R Other licence 107 forks Created Nov 2018 View on GitHub

Stars
815
47% of rgeirhos's stars
Rank
#80,468
of 851,984 repos, top 10%
In R
#133
of 4,064 repos, top 4%

Among 851,984 repos

Compare with

630–1k stars: 35,599 repos, including texture-vs-shape

Stars along the bottom, and repos in each band on a log scale

Quick stats

Next rank
+1 star to reach #80,358
Next milestone
+6,403 stars to reach the top 1%
Popular
One of 490,650 repos with 100+ stars
Yearly average
102 stars a year since Nov 2018

Stars counted Recounted daily Next count around