MTKD / README.md
nielsr's picture
nielsr HF Staff
Add/improve model card
d4f2ca4 verified
|
raw
history blame
2.03 kB
metadata
license: mit
library_name: pytorch
pipeline_tag: image-to-image

JL1-CD: A New Benchmark for Remote Sensing Change Detection and a Robust Multi-Teacher Knowledge Distillation Framework

This repository contains the model presented in the paper JL1-CD: A New Benchmark for Remote Sensing Change Detection and a Robust Multi-Teacher Knowledge Distillation Framework. This model utilizes a multi-teacher knowledge distillation (MTKD) framework for change detection (CD) in remote sensing images.

Dataset

The JL1-CD dataset is publicly available and can be downloaded from:

Usage

Install

To set up the environment, follow the installation instructions provided in the OpenCD repository.

Training

The training process for the MTKD framework consists of three steps. Below, we use the Changer-MiT-b0 model as an example:

Step 1: Train the original model

Step 2: Train teacher models for different CAR partitions (e.g., 3 partitions)

Step 3: Train the student model

Testing

Checkpoints

You can download checkpoint files from:

Citation

If you find the JL1-CD dataset or our work useful in your research, please consider citing our paper:

@article{liu2025jl1,
  title={JL1-CD: A New Benchmark for Remote Sensing Change Detection and a Robust Multi-Teacher Knowledge Distillation Framework},
  author={Liu, Ziyuan and Zhu, Ruifei and Gao, Long and Zhou, Yuanxiu and Ma, Jingyu and Gu, Yuantao},
  journal={arXiv preprint arXiv:2502.13407},
  year={2025}
}

Code: https://github.com/circleLZY/MTKD-CD.