论文标题
小波特征图像图像到图像的CNNS
Wavelet Feature Maps Compression for Image-to-Image CNNs
论文作者
论文摘要
卷积神经网络（CNN）以需要大量的计算资源而闻名，量化是压缩它们的最佳，最常见的方法之一。虽然积极的量化（即小于4位）在分类方面表现良好，但在图像到图像任务（例如语义分割和深度估计）中可能会导致严重的性能下降。在本文中，我们提出了小波压缩卷积（WCC） - 一种与点卷积集成的高分辨率激活图压缩方法的新方法，这是现代体系结构的主要计算成本。为此，我们使用有效且适合硬件的HAAR小波变换，以其在图像压缩方面的有效性而闻名，并在压缩激活图上定义了卷积。我们尝试从高分辨率输入中受益的各种任务。通过将WCC与光量化相结合，我们实现了相当于1-4位激活量化的压缩率，并且在性能中相对较小且更优雅的降解。我们的代码可在https://github.com/bgucompsci/waveletcompressedconvolution上找到。
Convolutional Neural Networks (CNNs) are known for requiring extensive computational resources, and quantization is among the best and most common methods for compressing them. While aggressive quantization (i.e., less than 4-bits) performs well for classification, it may cause severe performance degradation in image-to-image tasks such as semantic segmentation and depth estimation. In this paper, we propose Wavelet Compressed Convolution (WCC) -- a novel approach for high-resolution activation maps compression integrated with point-wise convolutions, which are the main computational cost of modern architectures. To this end, we use an efficient and hardware-friendly Haar-wavelet transform, known for its effectiveness in image compression, and define the convolution on the compressed activation map. We experiment with various tasks that benefit from high-resolution input. By combining WCC with light quantization, we achieve compression rates equivalent to 1-4bit activation quantization with relatively small and much more graceful degradation in performance. Our code is available at https://github.com/BGUCompSci/WaveletCompressedConvolution.
