Aug 13, 2022 · We consider the problem of active domain adaptation (ADA) to unlabeled target data, of which subset is actively selected and labeled given a budget constraint.

We consider the problem of active domain adaptation (ADA) to unlabeled target data, of which subset is actively selected and labeled given a budget constraint.

This is official code for "Combating Label Distribution Shift for Active Domain Adaptation" accepted in ECCV2022 - sehyun03/ADA-label-distribution-matching.

We consider the problem of active domain adaptation (ADA) to unlabeled target data, of which subset is actively selected and labeled given a budget constraint.

This work proposes a novel sampling strategy, which seeks target data that best approximate the entire target distribution as well as being representative, ...

Oct 23, 2022 · At its heart lies a novel sampling strategy, which seeks target data that best approximate the entire target distribution as well as being ...

Every component in LAMDA generally improves the performance of each domain adaptation scenario. Com- paring the second and the third rows of Table a2, one can ...

@inproceedings{hwang2022combating, title={Combating Label Distribution Shift for Active Domain Adaptation}, author={Hwang, Sehyun and Lee, Sohyun and Kim, ...

2022 ; Combating label distribution shift for active domain adaptation. Sehyun Hwang, Sohyun Lee, Sungyeon Kim, Jungseul Ok, and Suha Kwak. European Conference ...

Combating label distribution shift for active domain adaptation. S Hwang, S Lee, S Kim, J Ok, S Kwak. European Conference on Computer Vision, 549-566, 2022. 18 ...