Dataset Groups Activity Stream VisDA dataset for UDA The VisDA dataset is a large-scale dataset for unsupervised domain adaptation, which consists of 152,397 synthetic images and 55,388 real-world images from the real world. BibTex: @dataset{X_Peng_and_B_Usman_and_N_Kaushik_and_J_Hoffman_and_D_Wang_and_K_Saenko_2024, abstract = {The VisDA dataset is a large-scale dataset for unsupervised domain adaptation, which consists of 152,397 synthetic images and 55,388 real-world images from the real world.}, author = {X. Peng and B. Usman and N. Kaushik and J. Hoffman and D. Wang and K. Saenko}, doi = {10.57702/3ofu3qoh}, institution = {No Organization}, keyword = {'Domain Adaptation', 'Image Classification', 'VisDA'}, month = {dec}, publisher = {TIB}, title = {VisDA dataset for UDA}, url = {https://service.tib.eu/ldmservice/dataset/visda-dataset-for-uda}, year = {2024} }
