This is an official implementaion of the paper "SFCN+" in Learning from Synthetic Data for Crowd Counting in the Wild. More detialed information of the paper is shown in the project homepage.
- Python 2.7
- Pytorch 0.4.0
- TensorboardX (pip)
- torchvision (pip)
- easydict (pip)
- pandas (pip)
- Download the original UCF-QNRF Dataset [Link: Dropbox ]
- Resize the images and the locations of key points.
- Generate the density maps by using the code.
- Generate the segmentation maps.
The pre-trained resSFCN on GCC and the processed QNRF dataset: [Link]
- Run the train.py:
python train.py
. - See the training outputs:
Tensorboard --logdir=exp --port=6006
.
Errors on test set:
Note: the blue line is the result of using pre-trained GCC Dataset, and the red is the result of using pre-trained ImageNet.
Pre-trained ImageNet | Pre-trained GCC | |
---|---|---|
epoch 1,6 | ||
epoch 11,16 | ||
epoch 379,380 |
Column 1: input image; Column 2: density map GT; Column 3: density map prediction.
We reproduce some classic networks (MCNN, CSRNet, SANet, etc.) and some solid baseline networks (AlexNet, VGG, ResNet, etc.) on GCC dataset. Welcome to visit this link. It is under development, we will release it as soon as possible.
If you find this project useful for your research, please cite:
@inproceedings{wang2019learning,
title={Learning from Synthetic Data for Crowd Counting in the Wild},
author={Wang, Qi and Gao, Junyu and Lin, Wei and Yuan, Yuan},
booktitle={Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
year={2019}
}