LCFCN - ECCV 2018

Where are the Blobs: Counting by Localization with Point Supervision

Make the segmentation model learn to count and localize objects by adding a single line of code. Instead of applying the cross-entropy loss on dense per-pixel labels, apply the lcfcn loss on point-level annotations.

Usage

pip install git+https://github.com/ElementAI/LCFCN

from lcfcn import lcfcn_loss

# compute an CxHxW logits mask using any segmentation model
logits = seg_model.forward(images)

# compute loss given 'points' as HxW mask (1 pixel label per object)
loss = lcfcn_loss.compute_loss(points=points, probs=logits.sigmoid())

loss.backward()

Predicted Object Locations

Experiments

1. Install dependencies

pip install -r requirements.txt

This command installs pydicom and the Haven library which helps in managing the experiments.

2. Download Datasets

Shanghai Dataset

https://www.dropbox.com/s/kqqkl0exfshsw8v/mcnn_shtechB_110.h5?dl=1&fbclid=IwAR2LxhLqdyfEQx6olkTIUy-chTMhOzqMLQKcPErSTmiM5LhmeBnnGltIqVY

Trancos Dataset

wget http://agamenon.tsc.uah.es/Personales/rlopez/data/trancos/TRANCOS_v3.tar.gz

3. Train and Validate

python trainval.py -e trancos -d <datadir> -sb <savedir_base> -r 1

<datadir> is where the dataset is located.
<savedir_base> is where the experiment weights and results will be saved.
-e trancos specifies the trancos training hyper-parameters defined in exp_configs.py.

4. View Results

3.1 Launch Jupyter from terminal

> jupyter nbextension enable --py widgetsnbextension --sys-prefix
> jupyter notebook

3.2 Run the following from a Jupyter cell

from haven import haven_jupyter as hj
from haven import haven_results as hr

# path to where the experiments got saved
savedir_base = <savedir_base>

# filter exps
filterby_list = [('dataset.name','trancos')]
# get experiments
rm = hr.ResultManager(savedir_base=savedir_base, 
                      filterby_list=filterby_list, 
                      verbose=0)
# dashboard variables
legend_list = ['model.base']
title_list = ['dataset', 'model']
y_metrics = ['val_mae']

# launch dashboard
hj.get_dashboard(rm, vars(), wide_display=True)

This script outputs the following dashboard

Citation

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

@inproceedings{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},
  booktitle={Proceedings of the European Conference on Computer Vision (ECCV)},
  pages={547--562},
  year={2018}
}

sasakisasaki/LCFCN