LCFCN - ECCV 2018

Where are the Blobs: Counting by Localization with Point Supervision

Requirements

Pytorch version 0.4 or higher.

Description

Given a test image, the trained model outputs blobs in the image, then counts the number of predicted blobs (see Figure below).

Test on single image

We test a trained ResNet on a Trancos example image as follows:

python main.py -image_path figures/test.png \
                -model_path checkpoints/best_model_trancos_ResFCN.pth \
                -model_name ResFCN

The expected output is shown below, and the output image will be saved in the same directory as the test image.

Trancos test image	Trancos predicted image

Running the saved models

Download the checkpoints,

bash checkpoints/download.sh

Output the saved results,

python main.py -m summary -e trancos

Re-evaluate the saved model,

python main.py -m test -e trancos

Training the models from scratch

To train the model,

python main.py -m train -e trancos

Benchmark

Method	Trancos	Pascal
ResFCN	3.39	0.31
Paper	3.32	0.31

Citation

If you find the code useful for your research, please cite:

@Article{laradji2018blobs,
    title={Where are the Blobs: Counting by Localization with Point Supervision},
    author={Laradji, Issam H and Rostamzadeh, Negar and Pinheiro, Pedro O and Vazquez, David and Schmidt, Mark},
    journal = {ECCV},
    year = {2018}
}

zyweven/LCFCN