2PCNet: Two-Phase Consistency Training for Day-to-Night Unsupervised Domain Adaptive Object Detection

<img src="pytorch-logo-dark.png" width="10%">

This repo is the official implementation of our paper: <br> 2PCNet: Two-Phase Consistency Training for Day-to-Night Unsupervised Domain Adaptive Object Detection<br> Mikhail Kennerley, Jian-Gang Wang, Bharadwaj Veeravalli, Robby T. Tan <br> IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023 <br> [Paper]

<p align="center"> <img src="arch.jpg" width="95%"> </p>

Installation

Prerequisites

• Python ≥ 3.6
• PyTorch ≥ 1.5 and torchvision that matches the PyTorch installation.
• Detectron2 == 0.6

Dataset download

1. Download the datasets (BDD100K / SHIFT)

2. Split BDD100K and SHIFT into day and night labels using dataset information. Convert BDD100K and SHIFT labels to coco format. Alternatively, you can download our split (https://www.dropbox.com/scl/fo/258uzp6i0dz17zsj234r6/h?dl=0&rlkey=kb6brfk1oqc1ddsa3ulz8v9ei).

3. Organize the dataset with the following format

2pcnet/
└── datasets/
    └── bdd100k/
        ├── train/ 
            ├── img00001.jpg
            ├──...
        ├── val/ 
            ├── img00003.jpg
            ├──...
        ├── train_day.json
        ├── train_night.json
        ├── val_night.json
    └── shift/
        ├── train/ 
            ├── folder1
            ├──...
        ├── val/ 
            ├── folder1
            ├──...
        ├── train_day.json
        ├── train_night.json
        ├── val_night.json

    

Training

python train_net.py \
      --num-gpus 4 \
      --config configs/faster_rcnn_R50_bdd100k.yaml\
      OUTPUT_DIR output/bdd100k

Resume the training

python train_net.py \
      --resume \
      --num-gpus 4 \
      --config configs/faster_rcnn_R50_bdd100k.yaml MODEL.WEIGHTS <your weight>.pth

Evaluation

python train_net.py \
      --eval-only \
      --config configs/faster_rcnn_R50_bdd100k.yaml \
      MODEL.WEIGHTS <your weight>.pth

Pretrained Weights

| Dataset | Model Link | |-------------|----------------------------------------------------------------| | BDD100K | https://www.dropbox.com/s/812l6wdbonabp9k/model_final.pth?dl=0 | | SHIFT | Coming soon... |

Citation

If you use 2PCNet in your research or wish to refer to the results published in our paper, please use the following BibTeX entry:

@inproceedings{kennerley2023tpcnet,
  title={2PCNet: Two-Phase Consistency Training for Day-to-Night Unsupervised Domain Adaptive Object Detection},
  author={Mikhail Kennerley, Jian-Gang Wang, Bharadwaj Veeravalli, Robby T. Tan},
  booktitle={2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2023},
}

Acknowledgements

Code is adapted from Detectron2 and Adaptive Teacher.