데이터셋 탐색

DomainNet is a dataset of common objects in six different domain. All domains include 345 categories (classes) of objects such as Bracelet, plane, bird and cello. The domains include clipart: collection of clipart images; real: photos and real world images; sketch: sketches of specific objects; infograph: infographic images with specific object; painting artistic depictions of objects in the form of paintings and quickdraw: drawings of the worldwide players of game “Quick Draw!”. Source: What is being transferred in transfer learning? Image Source: http://ai.bu.edu/M3SDA/

Different from the setting of domain adaptation which uses all labeled source and unlabeled target domain examples for training, domain examples should be divided into two disjoint parts: training and test. UDE-DomainNet is built from DomainNet, so the performance of domain-adapted or domain-expanded models on the source and target domain can be evaluated.