Bamboo: Building Mega-Scale Vision Dataset Continually with Human-Machine Synergy

Open paper
Benchmark Model Rank Results
image-classification-on-food-101Bamboo (ViTB/16)#1Accuracy (%): 92.9
fine-grained-image-classification-on-sun397Bamboo (ViT-B/16)#3Accuracy: 79.5
image-classification-on-omnibenchmarkBamboo-R50#3Average Top-1 Accuracy: 45.4
image-classification-on-dtdBamboo (ViT-B/16)#4Accuracy: 81.9
image-classification-on-flowers-102Bamboo (ViT-B/16)#4Accuracy: 99.7
fine-grained-image-classification-on-oxford-iiit-petBamboo (ViT-B/16)#6Accuracy: 95.1%
fine-grained-image-classification-on-caltech-101Bamboo (ViT-B/16)#14Accuracy: 94.8
image-classification-on-objectnetVit B/16 (Bamboo)#32Top-1 Accuracy: 53.9
image-classification-on-objectnetResNet-50 (Bamboo)#51Top-1 Accuracy: 38.8
image-classification-on-cifar-100Bamboo (ViT-B/16)#41Percentage correct: 90.2
fine-grained-image-classification-on-stanford-carsBamboo (ViT-B/16)#42Accuracy: 93.9%
image-classification-on-cifar-10Bamboo (ViT-B/16)#68Percentage correct: 98.2

Rank counts only results with a code link.