Convolutional Neural Networks for Enhanced Compression Techniques
Matthew Gnacek, Cory Heatwole, David Fan, Marc P. Hoffman · 2021
This paper explores a compression framework combining convolutional neural networks (CNNs) with traditional image codecs. Various activation functions, codecs, optimizers, and upscaling techniques are tested to obtain the best image quality. While it was originally applied to grayscale images, the framework's performance with 8-bit IR images is also explored.