This repo is the official implementation of “Instant-Teaching: An End-to-End Semi-Supervised Object Detection Framework”.

  title={Instant-Teaching: An End-to-End Semi-Supervised Object Detection Framework},
  author={Zhou, Qiang and Yu, Chaohui and Wang, Zhibin and Qian, Qi and Li, Hao},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},

The code is based on MMdetection Toolbox.

1. Install

conda create -n instant_teaching python=3.7 -y
conda activate instant_teaching
conda install pytorch=1.7.1 torchvision cudatoolkit=10.1 -c pytorch
pip install mmcv==1.2.4

git clone
cd InstantTeaching
pip install -v -e .

2. Prepare COCO Dataset for Semi-Supervised Learning

mkdir -p datasets/coco
ln -s [coco_2017_downloaded] datasets/coco/coco_2017

python projects/InstantTeaching/tools/ datasets/coco/coco_2017/annotations/instances_train2017.json 10

Two files of instances_train2017_10_labeled.json and instances_train2017_90_unlabeled.json will be generated.

3. Training

CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python -m torch.distributed.launch \
    --nproc_per_node=8 --master_port=1234 tools/ \
    projects/InstantTeaching/configs/ \
    --launcher pytorch \
    --work-dir ${output_dir} \


View Github