论文标题
基于离散余弦变换的谐波卷积网络
Harmonic Convolutional Networks based on Discrete Cosine Transform
论文作者
论文摘要
卷积神经网络（CNN）学习过滤器，以捕获特征空间中的局部相关模式。我们建议学习这些过滤器作为由离散余弦变换（DCT）定义的预设光谱过滤器的组合。我们提出的基于DCT的谐波块取代了传统的卷积层，以生成新的或现有的CNN体系结构的部分或完全谐波版本。使用DCT能量压实属性，我们证明了如何通过频谱域中的冗余在谐波块中截断高频信息来有效地压缩谐波网络。我们报告了广泛的实验验证，这些验证表明在图像分类，对象检测和语义分割应用中将谐波块引入最先进的CNN模型的好处。
Convolutional neural networks (CNNs) learn filters in order to capture local correlation patterns in feature space. We propose to learn these filters as combinations of preset spectral filters defined by the Discrete Cosine Transform (DCT). Our proposed DCT-based harmonic blocks replace conventional convolutional layers to produce partially or fully harmonic versions of new or existing CNN architectures. Using DCT energy compaction properties, we demonstrate how the harmonic networks can be efficiently compressed by truncating high-frequency information in harmonic blocks thanks to the redundancies in the spectral domain. We report extensive experimental validation demonstrating benefits of the introduction of harmonic blocks into state-of-the-art CNN models in image classification, object detection and semantic segmentation applications.
