Table of Contents
  • Docs >
  • Train & Test >
  • Semi-automatic Object Detection Annotation with MMRotate and Label-Studio
Shortcuts

Semi-automatic Object Detection Annotation with MMRotate and Label-Studio

Annotation data is a time-consuming and laborious task. This article introduces how to perform semi-automatic annotation using the RTMDet-R algorithm in MMRotate in conjunction with Label-Studio software. Specifically, using RTMDet-R to predict image annotations and then refining the annotations with Label-Studio. Community users can refer to this process and methodology and apply it to other fields.

  • RTMDet-R: RTMDet-R is a high-precision single-stage object detection algorithm developed by OpenMMLab, open-sourced in the MMRotate rotated object detection toolbox. Its open-source license is Apache 2.0, and it can be used freely without restrictions by industrial users.

  • Label Studio is an excellent annotation software covering the functionality of dataset annotation in areas such as image classification, object detection, and segmentation.

In this article, we will use DOTA images for semi-automatic annotation.

Environment Configuration

To begin with, you need to create a virtual environment and then install PyTorch and MMCV. In this article, we will specify the versions of PyTorch and MMCV. Next, you can install MMDetection, MMRotate, Label-Studio, and label-studio-ml-backend using the following steps:

Create a virtual environment:

conda create -n rtmdet python=3.9 -y
conda activate rtmdet

Install PyTorch:

# Linux and Windows CPU only
pip install torch==1.10.1+cpu torchvision==0.11.2+cpu torchaudio==0.10.1 -f https://download.pytorch.org/whl/cpu/torch_stable.html
# Linux and Windows CUDA 11.3
pip install torch==1.10.1+cu113 torchvision==0.11.2+cu113 torchaudio==0.10.1 -f https://download.pytorch.org/whl/cu113/torch_stable.html
# OSX
pip install torch==1.10.1 torchvision==0.11.2 torchaudio==0.10.1

Install MMCV:

pip install -U openmim
mim install "mmcv>=2.0.0rc0"
# Installing mmcv will automatically install mmengine

Install MMDetection:

git clone https://github.com/vbti-development/onedl-mmdetection -b dev-3.x
cd mmdetection
pip install -v -e .

Install MMRotate:

git clone https://github.com/vbti-development/onedl-mmrotate/ -b dev-1.x
cd mmrotate
pip install -v -e .

Install Label-Studio and label-studio-ml-backend:

# Installing Label-Studio may take some time, if the version is not found, please use the official source
pip install label-studio==1.7.2
pip install label-studio-ml==1.0.9

Download the RTMDet-R weights:

cd path/to/mmrotate
mkdir work_dirs
cd work_dirs
wget https://mmassets.onedl.ai/mmrotate/v1.0/rotated_rtmdet/rotated_rtmdet_m-3x-dota/rotated_rtmdet_m-3x-dota-beeadda6.pth

Start the Service

Start the RTMDet-R backend inference service:

cd path/to/mmrotate
label-studio-ml start projects/LabelStudio/backend_template --with \
config_file=configs/rotated_rtmdet/rotated_rtmdet_m-3x-dota.py \
checkpoint_file=./work_dirs/rotated_rtmdet_m-3x-dota-beeadda6.pth \
device=cpu \
--port 8003
# Set device=cpu to use CPU inference, and replace cpu with cuda:0 to use GPU inference.

The RTMDet-R backend inference service has now been started. To configure it in the Label-Studio web system, use http://localhost:8003 as the backend inference service.

Now, start the Label-Studio web service:

label-studio start

Open your web browser and go to http://localhost:8080/ to see the Label-Studio interface.

Register a user and then create an RTMDet-R-Semiautomatic-Label project.

Download the example DOTA images by following MMRotate Preparing DOTA Dataset and import them using the Data Import button.

Then, select the Object Detection With Bounding Boxes template.

plane
ship
storage-tank
baseball-diamond
tennis-court
basketball-court
ground-track-field
harbor
bridge
large-vehicle
small-vehicle
helicopter
roundabout
soccer-ball-field
swimming-pool

Then, copy and add the above categories to Label-Studio and click Save.

In the Settings, click Add Model to add the RTMDet-R backend inference service.

Click Validate and Save, and then click Start Labeling.

If you see Connected as shown below, the backend inference service has been successfully added.

Start Semi-Automatic Labeling

Click on Label to start labeling.

We can see that the RTMDet-R backend inference service has successfully returned the predicted results and displayed them on the image. However, we noticed that there are some missed predicted bounding boxes.

We manually add some annotation boxes, adjust the position of the boxes to get the following corrected annotation, and then click Submit to complete the annotation of this image.

After submitting all images, click export to export the labeled dataset in COCO format.

Use VSCode to open the unzipped folder to see the labeled dataset, which includes the images and the annotation files in JSON format.

At this point, the semi-automatic labeling is complete. We can use this dataset to train a more accurate model in MMRotate and then continue semi-automatic labeling on newly collected images with this model. This way, we can iteratively expand the high-quality dataset and improve the accuracy of the model.