Harm-Net
2000년 도입 · 논문 1편에서 사용
A Harmonic Network, or Harm-Net, is a type of convolutional neural network that replaces convolutional layers with "harmonic blocks" that use Discrete Cosine Transform (DCT) filters. These blocks can be useful in truncating high-frequency information (possible due to the redundancies in the spectral domain).
출처: Harmonic Convolutional Networks based on Discrete Cosine Transform
소개 논문: Harmonic Convolutional Networks based on Discrete Cosine Transform
Convolutional Neural Networks · Computer Vision