Self-Supervised Real-to-Sim Scene Generation
        Aayush Prakash       Shoubhik Debnath    Jean-Francois Lafleche
      Eric Cameracci       Gavriel State  Stan Birchfield   Marc T. Law
                            NVIDIA

            Abstract
  Synthetic data is emerging as a promising solution
to the scalability issue of supervised deep learning, espe-
cially when real data are difficult to acquire or hard to
annotate. Synthetic data generation, however, can itself
be prohibitively expensive ...