Abstract: This paper presents Bottom-up Residual vector quantization for learned Image Compression (BRIC). This novel deep learning-based image compression method quantizes latent representations ...
Animation of validation reconstruction after each training epoch on the celebA-HQ 256x256 dataset. Using VIT encoder and decoder, and latent resolution of 32x32 This repository contains my PyTorch ...
Abstract: An important high-precision building vector mapping method automatically delineates building polygons from high-resolution remote sensing images. Deep learning methods have greatly improved ...
We’d like to note that after finishing our project and submitting the paper, we came across NegPiP. It’s a great community project that uses an approach very similar to ours. Although our work was ...