CACViT

Reported on 4 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 Vision4 results

Object CountingonFSC147
MAE(test)· 2023-05-08
9.13
best: 5.74 (CountGD)
SOTA
Vision Transformer Off-the-Shelf: A Surprising Baseline for Few-Shot Class-Agnostic Counting arXiv:2305.04440
Object CountingonFSC147
RMSE(test)· 2023-05-08
48.96
best: 24.09 (CountGD)
SOTA
Vision Transformer Off-the-Shelf: A Surprising Baseline for Few-Shot Class-Agnostic Counting arXiv:2305.04440
Object CountingonFSC147
MAE(val)· 2023-05-08
10.63
best: 7.1 (CountGD)
Vision Transformer Off-the-Shelf: A Surprising Baseline for Few-Shot Class-Agnostic Counting arXiv:2305.04440
Object CountingonFSC147
RMSE(val)· 2023-05-08
37.95
best: 26.08 (CountGD)
Vision Transformer Off-the-Shelf: A Surprising Baseline for Few-Shot Class-Agnostic Counting arXiv:2305.04440