This repository implements the classic ‘A Neural Algorithm of Artistic Style’ paper by Gatys et al., which uses convolutional neural networks to separate and recombine content and style from images. The implementation uses VGG-19 and Inception architectures to optimize pixel values through gradient descent, minimizing a content loss and style loss (Gram matrix based) to produce artistic images. It requires CUDA for GPU acceleration and includes model downloading scripts.