Home

Awesome

ShadowMaskFormer

A transformer-based approach for image shadow removal. This repository includes code for the following paper:

ShadowMaskFormer: Mask Augmented Patch Embedding for Shadow Removal

Training Environment

We test the code on PyTorch 1.10.2 + CUDA 11.3 + cuDNN 8.2.0.

  1. Create a new conda environment
conda create -n shadowmaskformer python=3.7
conda activate shadowmaskformer
  1. Install dependencies
conda install pytorch=1.10.2 torchvision torchaudio cudatoolkit=11.3 -c pytorch
pip install -r requirements.txt

Instructions

Please execute the following instructions to configure the parameters for running the program:

  1. Model Training python train.py --model (model name) --dataset (dataset name) --exp (exp name) e.g.: python train.py --model shadowmaskformer-b --dataset ISTD --exp istd

  2. Model Testing python test.py --model (model name) --dataset (dataset name) --exp (exp name) e.g.: python test.py --model shadowmaskformer-b --dataset ISTD --exp istd

Citation

@misc{li2024shadowmaskformermaskaugmentedpatch,
      title={ShadowMaskFormer: Mask Augmented Patch Embeddings for Shadow Removal}, 
      author={Zhuohao Li and Guoyang Xie and Guannan Jiang and Zhichao Lu},
      year={2024},
      eprint={2404.18433},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2404.18433}, 
}