Unofficial BCNet with Faster RCNN meta-architecture.

• Link to the original repository with FCOS meta-architecture: https://github.com/lkeab/BCNet
• Link to the paper: https://arxiv.org/abs/2103.12340.

This repo is initialized from the detectron2 repository, which runs faster-rcnn as default. The added BCNet source code can be found at these commits: mask head, modeling, data.

Installation

conda create -n dt2 python=3.7 -y
source activate dt2 

conda install pytorch==1.10.0 torchvision==0.11.0 cudatoolkit=11.3 -c pytorch
pip install ninja yacs cython matplotlib tqdm
pip install opencv-python==4.4.0.40
pip install scikit-image
pip install timm==0.4.12

# coco api
git clone https://github.com/cocodataset/cocoapi.git
cd cocoapi/PythonAPI
python setup.py build_ext install

cd Faster-RCNN-BCNet/
python3 setup.py build develop

pip install fvcore==0.1.1.dev200512 # just in case

Dataset registration

Follow the scripts of the original repository to create the right annotation file for BCNet.

Register the data in detectron2/data/datasets/builtin.py. See register_kins() in this file for references.

Available configs

Available config for reproducing the paper results on the KINS dataset: kins config

Training

Custom parameters at scripts/all.sh and run it for training.