arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2104.12686cs.CVcs.LG
收藏

基于深度卷积高斯混合模型的图像建模

Image Modeling with Deep Convolutional Gaussian Mixture Models

Fulda University of Applied Sciences(富尔达应用科学大学)

Alexander Gepperth, Benedikt Pfülb

摘要:

在这项概念性工作中,我们提出了深度卷积高斯混合模型(DCGMixture):一种新的深度分层高斯混合模型(GCGMixture),特别适合于描述和生成图像。香草(即,平面)GSTOM需要非常大量的组件来很好地描述图像,导致长的训练时间和记忆问题。DCG通过多个GMM层的堆叠架构避免了这一点,这些层通过卷积和池化操作连接。这允许以与深度CNN类似的方式利用图像的组合性。DCG算法可以通过随机梯度下降进行端到端的训练。这使它们与通过期望最大化训练的vanilla Gestro不同,后者需要事先进行k均值初始化,这在分层结构中是不可行的。为了使用DCGtron生成清晰的图像,我们引入了一种新的基于梯度的技术,通过卷积和池化等不可逆操作进行采样。基于MNIST和FashionMNIST数据集,我们通过展示其在聚类、采样和离群值检测方面优于平坦Gestrian来验证DCGestrian模型。

英文摘要:

In this conceptual work, we present Deep Convolutional Gaussian Mixture Models (DCGMMs): a new formulation of deep hierarchical Gaussian Mixture Models (GMMs) that is particularly suitable for describing and generating images. Vanilla (i.e., flat) GMMs require a very large number of components to describe images well, leading to long training times and memory issues. DCGMMs avoid this by a stacked architecture of multiple GMM layers, linked by convolution and pooling operations. This allows to exploit the compositionality of images in a similar way as deep CNNs do. DCGMMs can be trained end-to-end by Stochastic Gradient Descent. This sets them apart from vanilla GMMs which are trained by Expectation-Maximization, requiring a prior k-means initialization which is infeasible in a layered structure. For generating sharp images with DCGMMs, we introduce a new gradient-based technique for sampling through non-invertible operations like convolution and pooling. Based on the MNIST and FashionMNIST datasets, we validate the DCGMMs model by demonstrating its superiority over flat GMMs for clustering, sampling and outlier detection.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑