ML_GCN.pytorch
PyTorch implementation of Multi-Label Image Recognition with Graph Convolutional Networks, CVPR 2019.
Requirements
Please, install the following packages
- numpy
- torch-0.3.1
- torchnet
- torchvision-0.2.0
- tqdm
Download pretrain models
checkpoint/coco (GoogleDrive)
checkpoint/voc (GoogleDrive)
Options
lr
: learning ratelrp
: factor for learning rate of pretrained layers. The learning rate of the pretrained layers islr * lrp
batch-size
: number of images per batchimage-size
: size of the imageepochs
: number of training epochsevaluate
: evaluate model on validation setresume
: path to checkpoint
Demo VOC 2007
python3 demo_voc2007_gcn.py data/voc --image-size 448 --batch-size 32 -e --resume checkpoint/voc/voc_checkpoint.pth.tar
Demo COCO 2014
python3 demo_coco_gcn.py data/coco --image-size 448 --batch-size 32 -e --resume checkpoint/coco/coco_checkpoint.pth.tar
Citing this repository
If you find this code useful in your research, please consider citing us:
@inproceedings{ML_GCN_CVPR_2019,
author = {Zhao-Min, Chen and Xiu-Shen, Wei and Peng, Wang and Yanwen, Guo},
title = {{Multi-Label Image Recognition with Graph Convolutional Networks}},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
year = {2019}
}
Reference
This project is based on https://github.com/durandtibo/wildcat.pytorch