논문 함께 읽기/2D Image Classification (IC2D)
[IC2D] Drop an Octave: Reducing Spatial Redundancy in Convolution Neural Networks with Octave Convolution (ICCV2019)
안녕하세요. 지난 포스팅의 [IC2D] Searching for MobileNetV3 (ICCV2019)에서는 MobileNetV2와 MNAS + NetAdapt 알고리즘을 결합하여 좀 더 효율적인 모델인 MobileNetV3를 제안하였습니다. 오늘은 새로운 합성곱 연산인 OctConv에 대해서 소개하고자 합니다. Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave ConvolutionIn natural images, information is conveyed at different frequencies where higher frequencies are usually encoded with f..
