https://github.com/elerac/fullscreen
Display image in fullscreen for the projector-camera system.
https://github.com/elerac/fullscreen
display fullscreen imshow projector
Last synced: about 1 year ago
JSON representation
Display image in fullscreen for the projector-camera system.
- Host: GitHub
- URL: https://github.com/elerac/fullscreen
- Owner: elerac
- License: mit
- Created: 2019-11-28T09:20:29.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2024-12-02T06:07:18.000Z (over 1 year ago)
- Last Synced: 2025-03-24T13:21:28.154Z (about 1 year ago)
- Topics: display, fullscreen, imshow, projector
- Language: Python
- Homepage:
- Size: 683 KB
- Stars: 14
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# FullScreen
Display image in fullscreen for the projector-camera system.
Python has some GUI tools, but only a few of them are suitable for fullscreen display. In this repository, *Tkinter* or *OpenCV* or *PyQt5* can be used as the backend. The best backend depends on the environment (i.e., OS, monitor, version of the software), so choose the best one for you.
> [!NOTE]
> Recently, I released a new fullscreen program using OpenGL/GLFW in Python. I confirmed it could run at 60 fps, which achieves much faster pro-cam synchronization. You can check it out at https://github.com/elerac/pyglimshow.
## Requirement
### Tkinter backend (default)
- Pillow (PIL)
### OpenCV backend (optional)
- OpenCV
- screeninfo (from https://github.com/rr-/screeninfo)
### PyQt5 backend (optional)
- Pillow (PIL) or OpenCV
- PyQt5
## Usage
### Basic imshow
```python
import time
import cv2
from fullscreen import FullScreen
screen = FullScreen()
image = cv2.imread("image/himeji_castle.jpg", 1)
screen.imshow(image)
time.sleep(3)
```
### Get the size of the screen
```python
height, width, ch = screen.shape
```
### Choosing backend
```python
screen = FullScreen(backend="tkinter") # default
```
```python
screen = FullScreen(backend="cv2")
```
```python
screen = FullScreen(backend="PyQt5")
```
## Testing environment
| | OS | Hardware | Python | Backend |
| ---- | ---- | ---- | ---- | ---- |
| ✅ | macOS Catalina | MacBook Pro (13-inch, 2017) | 3.8 | Tkinter 8.6.11 |
| ✅ | macOS Catalina | MacBook Pro (13-inch, 2017) | 3.8 | OpenCV-Python 4.4 |
| ❌ | macOS Catalina | MacBook Pro (13-inch, 2017) | 3.8 | OpenCV-Python 4.5 |
| ✅ | macOS Catalina | MacBook Pro (13-inch, 2017) | 3.8 | PyQt5 5.15 |
| ❌ | macOS BigSur | MacBook Pro (13-inch, 2017) | 3.7 | Tkinter 8.5.9 |
| ✅ | macOS BigSur | MacBook Pro (13-inch, 2017) | 3.9 | Tkinter 8.6.11 |
| ❌ | macOS BigSur | MacBook Pro (13-inch, 2017) | 3.9 | OpenCV-Python 4.4 |
| ❌ | macOS BigSur | MacBook Pro (13-inch, 2017) | 3.9 | OpenCV-Python 4.5 |
| ❌ | macOS BigSur | Mac Pro (2019) | 3.9 | Tkinter 8.5.9 |
| ✅ | macOS BigSur | Mac Pro (2019) | 3.9 | Tkinter 8.6.11 |
| ✅ | macOS BigSur | Mac Pro (2019) | 3.8 | OpenCV-Python 4.4 |
| ❌ | macOS BigSur | Mac Pro (2019) | 3.8 | OpenCV-Python 4.5 |
| ✅ | Windows 10 | ThinkPad (L580) | 3.9 | Tkinter 8.6.9 |
| ✅ | Windows 10 | ThinkPad (L580) | 3.9 | OpenCV-Python 4.4 |
| ✅ | Windows 10 | ThinkPad (L580) | 3.9 | OpenCV-Python 4.5 |
| ✅ | Ubuntu 20 (GNOME) | | 3.8 | Tkinter 8.6.10 |
| ✅ | Ubuntu 20 (GNOME) | | 3.8 | OpenCV-Python 4.5 |
| 🤔 | Ubuntu 20 (KDE) | | 3.8 | Tkinter 8.8.10 |
| ✅ | Ubuntu 20 (KDE) | | 3.8 | OpenCV-Python 4.4 |
Although I have mainly tested on Mac, it should also work on Windows and Ubuntu as the software is cross-platform.
To summarize the above table, the following cases do not seem to work well ❌,
- Tkinter 8.5.9 does not work well on macOS. I think it is related to that [the Apple-supplied Tcl/Tk 8.5 has serious bugs](https://www.python.org/download/mac/tcltk/).
- OpenCV-Python 4.5 does not work well on macOS. I think it is related to that [Qt support has been dropped](https://github.com/opencv/opencv-python/releases/tag/48).
In some cases, it works but may not be perfect 🤔,
- On Ubuntu, Tkinter could not display the first image, but the second and later images work well.
-