Can I read Dynamic Distillation Network for Cross-Domain Few-Shot Recognition with Unlabeled Data on EtoBox?

Dynamic Distillation Network for Cross-Domain Few-Shot Recognition with Unlabeled Data by Islam, Ashraful; Chen, Chun-Fu; Panda, Rameswar; Karlinsky, Leonid; Feris, Rogerio; Radke, Richard J. is a scholarly article available to read on EtoBox.

What is Dynamic Distillation Network for Cross-Domain Few-Shot Recognition with Unlabeled Data about?

Most existing works in few-shot learning rely on meta-learning the network on a large base dataset which is typically from the same domain as the target dataset. We tackle the problem of cross-domain few-shot learning where there is a large shift between the base and target domain. The problem of cross-domain few-shot recognition with unlabeled target data is largely unaddressed in the literature. STARTUP was the first method that tackles this problem using self-training. However, it uses a fixed teacher pretrained on a labeled base dataset to create soft labels for the unlabeled target samples. As the base dataset and unlabeled dataset are from different domains, projecting the target images in the class-domain of the base dataset with a fixed pretrained model might be sub-optimal. We propose a simple dynamic distillation-based approach to facilitate unlabeled images from the novel/base dataset. We impose consistency regularization by calculating predictions from the weakly-augmented versions of the unlabeled images from a teacher network and matching it with the strongly augmented versions of the same images from a student network. The parameters of the teacher network are update

Author

Islam, Ashraful; Chen, Chun-Fu; Panda, Rameswar; Karlinsky, Leonid; Feris, Rogerio; Radke, Richard J.

Published

2021

Language

EN