Pinned Repositories
SCD-Net
[CVPR23] A cascaded diffusion captioning model with a novel semantic-conditional diffusion process that upgrades conventional diffusion model with additional semantic prior.
MG-Transformer
PKG-Transformer
RSIC-GMamba
SMART
[IEEE TPAMI 2024] This is the Pytorch code for our paper "SMART: Syntax-calibrated Multi-Aspect Relation Transformer for Change Captioning".