CVPR2021

DatasetGAN: Efficient Labeled Data Factory With Minimal Human Effort

Yuxuan Zhang, Huan Ling, Jun Gao, Kangxue Yin, Jean-Francois Lafleche, Adela Barriuso, Antonio Torralba, Sanja Fidler

摘要

Train DatasetGAN Manual annotation of few generated images Sample from the DatasetGAN a large synthetic dataset Train on synthetic and test on Real Images 2 3 4 1 Figure 1: DATASETGAN synthesizes image-annotation pairs, and can produce large high-quality datasets with detailed pixel-wise labels. Figure illustrates the 4 steps. (1 & 2). Leverage StyleGAN and annotate only a handful of synthesized images. Train a highly effective branch to generate labels. (3). Generate a huge synthetic dataset of annotated images authomatically. (4). Train your favorite approach with the synthetic dataset and test on real images.