Corpus ID: 182952301

Multi-objects Generation with Amortized Structural Regularization

@article{Xu2019MultiobjectsGW,
  title={Multi-objects Generation with Amortized Structural Regularization},
  author={Kun Xu and Chongxuan Li and Jun Zhu and B. Zhang},
  journal={ArXiv},
  year={2019},
  volume={abs/1906.03923}
}

• Kun Xu, Chongxuan Li, +1 author B. Zhang
• Published 2019
• Computer Science, Mathematics
• ArXiv

Deep generative models (DGMs) have shown promise in image generation. [...] Key Method In this paper, we propose the amortized structural regularization (ASR) framework, which adopts the posterior regularization (PR) to embed human knowledge into DGMs via a set of structural constraints. We derive a lower bound of the regularized log-likelihood, which can be jointly optimized with respect to the generative model and recognition model efficiently. Empirical results show that ASR significantly outperforms the DGM…Expand

View PDF on arXiv

Save to Library

Create Alert

Cite

Launch Research Feed