TKdata

This repository contains the dataset and pre-trained model weights associated with the paper "Research on the Semantic Segmentation of Thangka Images via an Improved PIDNet". It includes: Minimal anonymized thangka image dataset for semantic segmentation. Pre-trained weights of the improved PIDNet model proposed in the paper. Necessary files to replicate the experimental results in the study. All data and model weights are publicly available to support reproducibility and further research on cultural heritage image analysis and semantic segmentation. Related code repository: https://github.com/JikD-591/thangka-seg