https://github.com/ming71/yolov3-polygon
Arbitrary-oriented object detection based on yolov3.
https://github.com/ming71/yolov3-polygon
rotation-detector yolov3
Last synced: 26 days ago
JSON representation
Arbitrary-oriented object detection based on yolov3.
- Host: GitHub
- URL: https://github.com/ming71/yolov3-polygon
- Owner: ming71
- Created: 2020-10-21T13:56:24.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2020-12-18T04:26:28.000Z (over 4 years ago)
- Last Synced: 2024-10-28T06:00:25.940Z (6 months ago)
- Topics: rotation-detector, yolov3
- Language: Python
- Homepage:
- Size: 2.16 MB
- Stars: 134
- Watchers: 1
- Forks: 39
- Open Issues: 18
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- awesome-yolo-object-detection - ming71/yolov3-polygon - polygon?style=social"/> : Arbitrary-oriented object detection based on yolov3. (Object Detection Applications)
- awesome-yolo-object-detection - ming71/yolov3-polygon - polygon?style=social"/> : Arbitrary-oriented object detection based on yolov3. (Applications)
README
# Yolov3-Polygon
This is a implementation of rotation object detecion based on [YOLOv3-quadrangle](https://github.com/JKBox/YOLOv3-quadrangle). I upgraded it to support pytorch 1.1 or higher and fix some bugs. Object detection in arbitrary orientations is achieved by detecting four corner points, the model has been tested on remote sensing dataset UCAS-AOD. The results and trained models can be found [here](https://github.com/ming71/UCAS-AOD-benchmark).
### training
#### dataset
The annotations of your own dataset need to be converted into DOTA format.
#### imagesets
Generate imageset file via `utils/generate_imageset.py`
#### config
Run `utils/kmeans.py` to generate preset anchors.
Modify two parts in `cfg/yolov3.cfg` : 1. classes 2. conv filter before yolo layer should be `(8+cls+1)*3`.
Modify training sets in `.data` file.
Modify classnames in `data/*.names`.
### eval
prepare labels:
```
python datasets/UCAS_AOD/ucas_aod2gt.py
```
conduct evaluation:
```
python eval.py
```
### detect
```
python detect.py
```
### Detections
### Thanks to
[YOLOv3-quadrangle](https://github.com/JKBox/YOLOv3-quadrangle)
[mAP calculation](https://github.com/Cartucho/mAP)
[ultralytics yolov3](https://github.com/ultralytics/yolov3)