VisDA-2017

홈페이지 · 논문 223편

VisDA-2017 is a simulation-to-real dataset for domain adaptation with over 280,000 images across 12 categories in the training, validation and testing domains. The training images are generated from the same object under different circumstances, while the validation images are collected from MSCOCO.. Source: Gradually Vanishing Bridge for Adversarial Domain Adaptation Image Source: http://ai.bu.edu/visda-2017/

Images

벤치마크

Domain Adaptation on VisDA2017 결과 28개

Universal Domain Adaptation on VisDA2017 결과 26개

Unsupervised Domain Adaptation on VisDA2017 결과 26개

Source-Free Domain Adaptation on VisDA-2017 결과 20개

Partial Domain Adaptation on VisDA2017 결과 3개

Unsupervised Domain Adaptation on VisDA-2017 결과 2개