open-mmlab / mmdetection

Open MMLab Detection Toolbox



mmdetection is an open source object detection toolbox based on PyTorch. It is a part of the open-mmlab project developed by Multimedia Laboratory, CUHK.

Major features

  • Modular Design

    One can easily construct a customized object detection framework by combining different components.

  • Support of multiple frameworks out of box

    The toolbox directly supports popular detection frameworks, e.g. Faster RCNN, Mask RCNN, RetinaNet, etc.

  • Efficient

    All basic bbox and mask operations run on GPUs now. The training speed is about 5% ~ 20% faster than Detectron for different models.

  • State of the art

    This was the codebase of the MMDet team, who won the COCO Detection 2018 challenge.

Apart from mmdetection, we also released a library mmcv for computer vision research, which is heavily depended on by this toolbox.


This project is released under the GPLv3 license.

Benchmark and model zoo

We provide our baseline results and the comparision with Detectron, the most popular detection projects. Results and models are available in the Model zoo.



  • Linux (tested on Ubuntu 16.04 and CentOS 7.2)
  • Python 3.4+
  • PyTorch 0.4.1 and torchvision
  • Cython
  • mmcv

Install mmdetection

a. Install PyTorch 0.4.1 and torchvision following the official instructions.

b. Clone the mmdetection repository.

git clone

c. Compile cuda extensions.

cd mmdetection
pip install cython  # or "conda install cython" if you prefer conda
./  # or "PYTHON=python3 ./" if you use system python3 without virtual environments

d. Install mmdetection (other dependencies will be installed automatically).

python(3) install  # add --user if you want to install it locally
# or "pip install ."

Note: You need to run the last step each time you pull updates from github. The git commit id will be written to the version number and also saved in trained models.

Prepare COCO dataset.

It is recommended to symlink the dataset root to $MMDETECTION/data.

├── mmdet
├── tools
├── configs
├── data
│   ├── coco
│   │   ├── annotations
│   │   ├── train2017
│   │   ├── val2017
│   │   ├── test2017

Here is a script for setting up mmdetection with conda for reference.

Inference with pretrained models

Test a dataset

  • single GPU testing
  • multiple GPU testing
  • visualize detection results

We allow to run one or multiple processes on each GPU, e.g. 8 processes on 8 GPU or 16 processes on 8 GPU. When the GPU workload is not very heavy for a single process, running multiple processes will accelerate the testing, which is specified with the argument --proc_per_gpu <PROCESS_NUM>.

To test a dataset and save the results.

python tools/ <CONFIG_FILE> <CHECKPOINT_FILE> --gpus <GPU_NUM> --out <OUT_FILE>

To perform evaluation after testing, add --eval <EVAL_TYPES>. Supported types are:

  • proposal_fast: eval recalls of proposals with our own codes. (supposed to get the same results as the official evaluation)
  • proposal: eval recalls of proposals with the official code provided by COCO.
  • bbox: eval box AP with the official code provided by COCO.
  • segm: eval mask AP with the official code provided by COCO.
  • keypoints: eval keypoint AP with the official code provided by COCO.

For example, to evaluate Mask R-CNN with 8 GPUs and save the result as results.pkl.

python tools/ configs/ <CHECKPOINT_FILE> --gpus 8 --out results.pkl --eval bbox segm

It is also convenient to visualize the results during testing by adding an argument --show.

python tools/ <CONFIG_FILE> <CHECKPOINT_FILE> --show

Test image(s)

We provide some high-level apis (experimental) to test an image.

import mmcv
from mmcv.runner import load_checkpoint
from mmdet.models import build_detector
from mmdet.apis import inference_detector, show_result

cfg = mmcv.Config.fromfile('configs/')
cfg.model.pretrained = None

# construct the model and load checkpoint
model = build_detector(cfg.model, test_cfg=cfg.test_cfg)
_ = load_checkpoint(model, '')

# test a single image
img = mmcv.imread('test.jpg')
result = inference_detector(model, img, cfg)
show_result(img, result)

# test a list of images
imgs = ['test1.jpg', 'test2.jpg']
for i, result in enumerate(inference_detector(model, imgs, cfg, device='cuda:0')):
    print(i, imgs[i])
    show_result(imgs[i], result)

Train a model

mmdetection implements distributed training and non-distributed training, which uses MMDistributedDataParallel and MMDataParallel respectively.

We suggest using distributed training even on a single machine, which is faster, and non-distributed training are left for debugging or other purposes.

Distributed training

mmdetection potentially supports multiple launch methods, e.g., PyTorch’s built-in launch utility, slurm and MPI.

We provide a training script using the launch utility provided by PyTorch.

./tools/ <CONFIG_FILE> <GPU_NUM> [optional arguments]

Supported arguments are:

  • --validate: perform evaluation every k (default=1) epochs during the training.
  • --work_dir <WORK_DIR>: if specified, the path in config file will be overwritten.

Non-distributed training

python tools/ <CONFIG_FILE> --gpus <GPU_NUM> --work_dir <WORK_DIR> --validate

Expected results in WORK_DIR:

  • log file
  • saved checkpoints (every k epochs, defaults=1)
  • a symbol link to the latest checkpoint

Technical details

Some implementation details and project structures are described in the technical details.

Currently unrated

Recent Posts






RSS / Atom