Initialization Using Perlin Noise for Training Networks with a Limited Amount of Data
This addresses the challenge of data scarcity in image classification, though it appears incremental as it builds on existing initialization techniques.
The paper tackles the problem of training image classification networks with limited data by proposing a novel initialization method using Perlin noise, which outperforms conventional methods on four datasets.
We propose a novel network initialization method using Perlin noise for training image classification networks with a limited amount of data. Our main idea is to initialize the network parameters by solving an artificial noise classification problem, where the aim is to classify Perlin noise samples into their noise categories. Specifically, the proposed method consists of two steps. First, it generates Perlin noise samples with category labels defined based on noise complexity. Second, it solves a classification problem, in which network parameters are optimized to classify the generated noise samples. This method produces a reasonable set of initial weights (filters) for image classification. To the best of our knowledge, this is the first work to initialize networks by solving an artificial optimization problem without using any real-world images. Our experiments show that the proposed method outperforms conventional initialization methods on four image classification datasets.