https://github.com/omar-abdelgawad/sperm-keypoint-detection
Trained YOLOv8 model on Pose estimation with sperm cells.
https://github.com/omar-abdelgawad/sperm-keypoint-detection
computer-vision keypoint-detection pose-estimation sperm-detection sperm-tracking yolov8 yolov8-pose yolov8x
Last synced: about 1 year ago
JSON representation
Trained YOLOv8 model on Pose estimation with sperm cells.
- Host: GitHub
- URL: https://github.com/omar-abdelgawad/sperm-keypoint-detection
- Owner: omar-abdelgawad
- License: mit
- Created: 2023-09-15T13:59:51.000Z (almost 3 years ago)
- Default Branch: master
- Last Pushed: 2024-07-17T23:54:17.000Z (about 2 years ago)
- Last Synced: 2024-11-14T07:46:30.068Z (almost 2 years ago)
- Topics: computer-vision, keypoint-detection, pose-estimation, sperm-detection, sperm-tracking, yolov8, yolov8-pose, yolov8x
- Language: Python
- Homepage: https://omar-abdelgawad.github.io/sperm-keypoint-detection/
- Size: 54.7 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# sperm-keypoint-detection
Trained YOLOv8 model on Pose estimation with sperm cells.
Note: The application comes in a "dist" folder containing one executable called "app.exe" and a "model" folder. It is advised to ignore everything in the "dist" folder except the executable file.
## Introduction
The first version of this application uses a command line interface. This means you have to run the executable with some arguments.
The following 3 are the current required arguments:
- "--input_path" or "-i" is the relative or absolute path of the input video.
- "--magnif" or "-m" is the magnification used for capturing the video.
- "--rate" or "-r" is the sampling rate for the input video. Ex: 736.
## Usage
In order to use the application, follow the following steps.
1. Navigate to the parent folder of the "dist" fodler.
1. press shift+right click on any empty space. This should open a pop-up window.
1. press "open in terminal". You should find a powershell terminal opened with the project path ready.
1. run
```bash
.\app.exe --input_path your/input/path --magnif your_magnif --rate your_fps
```
an example would be like the following.
```bash
.\app.exe --input_path '.\f16 736.79.avi' --magnif 40x --rate 736
```
You could also write a short version.
```bash
.\app.exe -i '.\f16 736.79.avi' -m 40x -r 736
```
1. The executable should run succesfully.
## Notes
1. The application will automatically look for gpu acces. If cuda is not installed it will use the cpu instead. However using a cpu could be significantly slower (About 20x slower).