w568w / SPDCN-CAC

BMVC-2022 paper "Scale-Prior Deformable Convolution for Class-Agnostic Counting"(https://bmvc2022.mpi-inf.mpg.de/313)

SPDCN

[Homepage][paper][Poster]

official code for BMVC-2022 paper "Scale-Prior Deformable Convolution for Exemplar-Guided Class-Agnostic Counting"

Requirement

We use Singularity to build the enviroment. Download our enviroment: excalibur.sif. If you'd like to create environement yourself, the following python packages are required:

pytorch == 1.9.0
torchvision == 0.10.0
mmcv == 1.3.13
timm == 0.4.12
termcolor
yacs
einops

Data Preparation

• Download FSC-147
• modify the root in line 12 of datasets/gendata384x576.py to the local path of FSC-147.
• running the file datasets/gendata384x576.py

Training

• modify the datapath in run.sh to the local path of FSC-147 dataset
• using singularity: singularity exec --bind --nv path_to_excalibur.sif ./run.sh
• using your own environment: ./run.sh

A training log is shown in md-files/training.log, and corresponding checkpoint is uploaded here.

Citation

@inproceedings{Lin_2022_BMVC,
author    = {Wei Lin and Kunlin Yang and Xinzhu Ma and Junyu Gao and Lingbo Liu and Shinan Liu and Jun Hou and Shuai Yi and Antoni Chan},
title     = {Scale-Prior Deformable Convolution for Exemplar-Guided Class-Agnostic Counting},
booktitle = {33rd British Machine Vision Conference 2022, {BMVC} 2022, London, UK, November 21-24, 2022},
publisher = {{BMVA} Press},
year      = {2022},
url       = {https://bmvc2022.mpi-inf.mpg.de/0313.pdf}
}