论文标题
Neurips 2019 DISENTANGREMT挑战：通过学习的卷积特征图的汇总改善了分解
NeurIPS 2019 Disentanglement Challenge: Improved Disentanglement through Learned Aggregation of Convolutional Feature Maps
论文作者
论文摘要
该报告向我们的第2阶段提交给Neurips 2019 Disentangrement Challenge提出了一种简单的图像预处理方法，用于学习解开潜在因素。我们建议在ImageNet数据库上预处理的网络中获得的区域汇总特征图训练变异自动编码器，并利用这些特征中包含的隐式归纳偏见以用于分离。可以通过对辅助任务上的特征图明确调整对挑战的特征图，例如角度，位置估计或颜色分类，从而进一步增强这种偏差。我们的方法在挑战的第2阶段获得了第二名。代码可从https://github.com/mseitzer/neurips2019-disentanglement-challenge获得。
This report to our stage 2 submission to the NeurIPS 2019 disentanglement challenge presents a simple image preprocessing method for learning disentangled latent factors. We propose to train a variational autoencoder on regionally aggregated feature maps obtained from networks pretrained on the ImageNet database, utilizing the implicit inductive bias contained in those features for disentanglement. This bias can be further enhanced by explicitly fine-tuning the feature maps on auxiliary tasks useful for the challenge, such as angle, position estimation, or color classification. Our approach achieved the 2nd place in stage 2 of the challenge. Code is available at https://github.com/mseitzer/neurips2019-disentanglement-challenge.
