Superpixel Positional Encoding to Improve ViT-based Semantic Segmentation Models

Open Webpage

Roberto Amoroso, Matteo Tomei, Lorenzo Baraldi, Rita Cucchiara

Published: 01 Jan 2023, Last Modified: 14 Nov 2024BMVC 2023EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Abstract