README.md

A-GCN

Learning Label Correlations for Multi-Label Image Recognition with Graph Networks

Requirements

* Pytorch 0.3.1
* Python 3.6

Demo COCO 2014

Please enter the main folder, and run

python3 demo_coco__adapt.py --data data/coco --image-size 448 --batch-size 32 -e --resume checkpoint/coco/coco_checkpoint.pth.tar

Our pretrained models

The pre-trained model on coco at GoogleDrive
The pre-trained model on fashion550k at GoogleDrive

Citing this repository

If you find our work helpful in your research, please kindly cite our paper:

Qing iI, Xiaojiang Peng, Yu Qiao, Qiang Peng. Learning label correlations for multi-label image recognition with graph networks[J]. Pattern Recognition Letters, 2020, 138: 378-384.

bib:

@article{li2020learning,
 title={Learning label correlations for multi-label image recognition with graph networks},
 author={Li, Qing and Peng, Xiaojiang and Qiao, Yu and Peng, Qiang},
 journal={Pattern Recognition Letters},
 volume={138},
 pages={378--384},
 year={2020},
 publisher={Elsevier}
 }

Reference

This project is based on https://github.com/Megvii-Nanjing/ML_GCN