Visual Fourier Prompt Tuning

Benchmark Model Rank Results
monocular-depth-estimation-on-clevr-distanceVFPT ViT-B/16 (VTAB-1k)–Accuracy: 63.2
object-counting-on-clevr-countVFPT ViT-B/16 (VTAB-1k)–Accuracy: 75.8