jeffaudi / SuperYOLO

SuperYOLO

SuperYOLO

This code will be completely released after our article is received！！！！

Please wait patiently！！！Thank you！！！

Requirements

pip install -r requirements.txt

Train

1. Prepare training data

• 1.1 In order to realize the SR assisted branch, the input images of the network are downsampled from 1024 x 1024 size to 512 x 512 during the training process. In the test process, the image size is 512 x 512, which is consistent with the input of other algorithms compared.

• 1.2 Download VEDAI data for our experiment from baiduyun (code: hvi4).

• 1.3 Note that we transform the labels of the dataset to be horizontal boxes by transform code. You shoud run transform.py before training the model.

2. Begin to train

python train_up.py --cfg models/SRyolo_noFocus.yaml --super --train_img_size 1024 --hr_input --data data/SRvedai.yaml --ch 4

Test

Pretrained Checkpoints

You can use our pretrained checkpoints for test process. Download pre-trained model here (code：lldz) and put it in here.

python test.py --weights runs/train/exp/best.pt --input_mode RGB+IR 

Results

Visualization of results

Citation

@misc{https://doi.org/10.48550/arxiv.2209.13351, doi = {10.48550/ARXIV.2209.13351},

url = {https://arxiv.org/abs/2209.13351},

author = {Zhang, Jiaqing and Lei, Jie and Xie, Weiying and Fang, Zhenman and Li, Yunsong and Du, Qian},

keywords = {Computer Vision and Pattern Recognition (cs.CV), FOS: Computer and information sciences, FOS: Computer and information sciences},

title = {SuperYOLO: Super Resolution Assisted Object Detection in Multimodal Remote Sensing Imagery},

publisher = {arXiv},

year = {2022},

copyright = {arXiv.org perpetual, non-exclusive license} }

Acknowledgements

This code is built on YOLOv5 (PyTorch). We thank the authors for sharing the codes.