https://github.com/zehuichen123/ddod

[ACM MM 2021] Disentangle your Dense Object Detector, Oral
https://github.com/zehuichen123/ddod

Last synced: about 1 month ago
JSON representation

[ACM MM 2021] Disentangle your Dense Object Detector, Oral

• Host: GitHub
• URL: https://github.com/zehuichen123/ddod
• Owner: zehuichen123
• License: apache-2.0
• Created: 2021-07-05T05:41:06.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2021-10-26T03:23:29.000Z (about 3 years ago)
• Last Synced: 2023-03-05T14:50:51.885Z (almost 2 years ago)
• Language: Python
• Homepage:
• Size: 18.7 MB
• Stars: 53
• Watchers: 2
• Forks: 7
• Open Issues: 4
• Metadata Files:
  • Readme: README.md
  • Contributing: .github/CONTRIBUTING.md
  • License: LICENSE
  • Code of conduct: .github/CODE_OF_CONDUCT.md

Awesome Lists containing this project

README

        
# Disentangle Your Dense Object Detector

This repo contains the supported code and configuration files to reproduce object detection results of [Disentangle Your Dense Object Detector](https://arxiv.org/abs/2107.02963). It is based on [mmdetection](https://github.com/open-mmlab/mmdetection).

## Results and Models

| Model | Backbone | Lr Schd | box mAP | AP50 | AP75 | APs | APm | APl |

| :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |:---: |

| ATSS(IoU) | ResNet50 | 1x | 39.4 | 56.6 | 42.6 | 23.9 | 42.5 | 49.6 |

| DDOD | ResNet50 | 1x | 41.6 | 59.9 | 45.2 | 23.9 | 44.9 | 54.4 |

| DDOD-FCOS | ResNet50 | 1x | 41.6 | 59.9 | 45.3 | 24.0 | 44.6 | 54.8 |

## Usage

### Installation

Please refer to [get_started.md](https://github.com/open-mmlab/mmdetection/blob/master/docs/get_started.md) for installation and dataset preparation.

### Inference

```

# multi-gpu testing

tools/dist_test.sh coco_cfg/ddod_r50_1x.py  8 --eval bbox

```

### Training

To train a detector with pre-trained models, run:

```

# multi-gpu training

tools/dist_train.sh coco_cfg/ddod_r50_1x.py 8

```

## Citing DDOD

```

@misc{chen2021disentangle,

      title={Disentangle Your Dense Object Detector}, 

      author={Zehui Chen and Chenhongyi Yang and Qiaofei Li and Feng Zhao and Zhengjun Zha and Feng Wu},

      year={2021},

      eprint={2107.02963},

      archivePrefix={arXiv},

      primaryClass={cs.CV}

}

```