RADAM (ConvNeXt-L)

Reported on 2 benchmarks across 1 task · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision2 results

Image ClassificationonDTD
Accuracy· uses extra data· 2023-03-08
84
best: 90 (Linear FT(ViT-L/14))
SOTA
RADAM: Texture Recognition through Randomized Aggregated Encoding of Deep Activation Maps arXiv:2303.04554
Image ClassificationonFMD (materials)
Accuracy (%)· uses extra data· 2023-03-08
95.2
SOTA
RADAM: Texture Recognition through Randomized Aggregated Encoding of Deep Activation Maps arXiv:2303.04554