Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

https://github.com/fpv-iplab/da-retinanet

Official Detectron2 implementation of DA-RetinaNet, An unsupervised domain adaptation scheme for single-stage artwork recognition in cultural sites, Image and Vision Computing (IMAVIS) 2021
https://github.com/fpv-iplab/da-retinanet

cityscapes cityscapes-dataset coco-annotations cyclegan da-retinanet detectron2 domain-adaptation domain-adaptation-object-detection foggy-cityscapes google-colab object-detection pascal-voc retinanet unsupervised-domain-adaptation

Last synced: 5 months ago
JSON representation

Official Detectron2 implementation of DA-RetinaNet, An unsupervised domain adaptation scheme for single-stage artwork recognition in cultural sites, Image and Vision Computing (IMAVIS) 2021

Awesome Lists containing this project

README 

          
# Detectron2 implementation of DA-RetinaNet

This is the implementation of our Image and Vision Computing 2021 work 'An unsupervised domain adaptation scheme for single-stage artwork recognition in cultural sites'. The aim is to reduce the gap between source and target distribution improving the object detector performance on the target domain when training and test data belong to different distributions. The original paper can be found [here](https://www.sciencedirect.com/science/article/pii/S0262885621000032?dgcid=rss_sd_all).
 If you want to use this code with your dataset, please follow the following guide. 



Please leave a star ⭐ and cite the following [paper](http://www.sciencedirect.com/science/article/pii/S0262885621000032) if you use this repository for your project.

```

@article{PASQUALINO2021104098,

    title = "An unsupervised domain adaptation scheme for single-stage artwork recognition in cultural sites",

    journal = "Image and Vision Computing",

    pages = "104098",

    year = "2021",

    issn = "0262-8856",

    doi = "https://doi.org/10.1016/j.imavis.2021.104098",

    author = "Giovanni Pasqualino and Antonino Furnari and Giovanni Signorello and Giovanni Maria Farinella",

}

```

## DA-RetinaNet Architecture




## Installation

You can use this repo following one of these three methods:


NB: Detectron2 0.6 is required, installing other versions this code will not work.



### Google Colab

Quickstart here 👉 [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/gist/GiovanniPasq/e8cc5e9d955ad01bf11df993ba276909/da-retinanet.ipynb)


Or load and run the ```DA-RetinaNet.ipynb``` on Google Colab following the instructions inside the notebook.



### Detectron2 on your PC

Follow the official guide to install [Detectron2 0.6](https://github.com/facebookresearch/detectron2/blob/master/INSTALL.md)


Or


Download the official Detectron2 0.6 from [here](https://github.com/facebookresearch/detectron2/archive/refs/tags/v0.6.zip)


Unzip the file and rename it in detectron2


run ```python -m pip install -e detectron2```



### Detectron2 via Dockerfile

Follow these instructions:

```

cd docker/

# Build 

docker build -t detectron2:v0 .



# Launch

docker run --gpus all -it --shm-size=8gb -v /home/yourpath/:/home/yourpath --name=name_container detectron2:v0

```

If you exit from the container you can restart it using:

```

docker start name_container

docker exec -it name_container /bin/bash

```



## Dataset

Create the Cityscapes-Foggy Cityscapes dataset following the instructions available [here](https://github.com/yuhuayc/da-faster-rcnn/tree/master/prepare_data)


The UDA-CH dataset is available [here](https://iplab.dmi.unict.it/EGO-CH-OBJ-UDA/EGO-CH-OBJ-UDA.zip)

## Data Preparation

If you want to use this code with your dataset arrange the dataset in the format of COCO or PASCAL VOC. 


For COCO annotations, inside the script ```uda_train.py``` register your dataset using: 



```register_coco_instances("dataset_name_source_training",{},"path_annotations","path_images")```


```register_coco_instances("dataset_name_target_training",{},"path_annotations","path_images")```


```register_coco_instances("dataset_name_target_test",{},"path_annotations","path_images")```



For PASCAL VOC annotations, inside the ```cityscape_train.py``` register your dataset using: 



```register_pascal_voc("city_trainS", "cityscape/VOC2007/", "train_s", 2007, ['car','person','rider','truck','bus','train','motorcycle','bicycle'])```


```register_pascal_voc("city_trainT", "cityscape/VOC2007/", "train_t", 2007, ['car','person','rider','truck','bus','train','motorcycle','bicycle'])```


```register_pascal_voc("city_testT", "cityscape/VOC2007/", "test_t", 2007, ['car','person','rider','truck','bus','train','motorcycle','bicycle'])```



You need to replace the parameters inside the ```register_pascal_voc()``` function according to your dataset name and classes. 



## Training

Replace at the following path ```detectron2/modeling/meta_arch/``` the ```dense_detector.py``` script with our ```dense_detector.py```. 


Do the same for the ```fpn.py``` file at the path ```detectron2/modeling/backbone/```



Run the script ```uda_train.py``` for COCO annotations or ```cityscape_train.py``` for PASCAL VOC annotations. 



Trained model on Cityscapes to FoggyCityscapes is available at this link: 


[DA-RetinaNet_Cityscapes](https://iplab.dmi.unict.it/EGO-CH-OBJ-UDA/DA-RetinaNet_Cityscapes.pth) 



Trained models on the proposed UDA-CH dataset are available at these links: 


[DA-RetinaNet](https://iplab.dmi.unict.it/EGO-CH-OBJ-UDA/DA-RetinaNet.pth) 


[DA-RetinaNet-CycleGAN](https://iplab.dmi.unict.it/EGO-CH-OBJ-UDA/DA-RetinaNet-CycleGAN.pth) 



## Testing

If you want to test the model load the new weights, set to 0 the number of iterations and rerun the same script used for the training.



## Results on Cityscapes -> Foggy Cityscapes



Results adaptation between Cityscapes and Foggy Cityscapes dataset. The performance scores of the methods marked with the “*” symbol are reported from the authors of their respective papers.




  

    Model

    mAP

  

  

    Faster RCNN*

    20.30%

  

  

    DA-Faster RCNN*

    27.60%

  

  

     StrongWeak*

    34.30%

  

  

    Diversify and Match*

    34.60%

  

  

    DA-RetinaNet

    44.87%

  

  

    RetinaNet (Oracle)

    53.46%

  



## Results on the proposed dataset Synthetic -> Real



  Results of DA-Faster RCNN, Strong-Weak and the proposed DA-RetinaNet combined with image-to-image translation approach.




  

    

    image to image translation (CycleGAN)

  

  

    Object Detector

    None

    Synthetic to Real

  

   

    DA-Faster RCNN

    12.94%

    33.20%

  

   

     StrongWeak

    25.12%

    47.70%

  

  

    DA-RetinaNet

    31.04%

    58.01%

  



## Other Works

[STMDA-RetinaNet](https://github.com/fpv-iplab/STMDA-RetinaNet)


[Detectron2 implementation of DA-Faster RCNN](https://github.com/GiovanniPasq/DA-Faster-RCNN)