The Classification of Multi-Domain Samples Based on the Cooperation of Multiple Models

Figure 6

The difference between the output of trained models. The ground truth is label 1 sample of domain 0. (a) The models trained on CIFAR-10. (b) The models trained on CIFAR-100. (c) The models trained on Mini-ImageNet. (d) The models trained on EuroSAT. (e) The models trained on Intel image classification.

(a)

(b)

(c)

(d)

(e)