update README.md to enhance camera support details and clarify features

saeugetier · saeugetier · commit 132dcc9fcd00 · 2026-01-11T10:08:04.000+01:00
diff --git a/README.md b/README.md
@@ -2,32 +2,37 @@
 
 ## Overview
 
-The goal of the project is to provide a configurable photobooth software for Raspberry Pi or a PC. 
+The goal of the project is to provide a configurable photobooth software for Raspberry Pi or a PC.
 
 Main features are:
 - User Interface with touchscreen input (input via mouse is not recommended). Control via buttons may be added in future.
-- Photo preview and capture via V4L2 (Raspberry Pi Camera or Webcam) ~~or via GPhoto2 supported cameras.~~ (currently GPhoto2 is not supported. Will be backported later.)
-- Printout with Canon Selphy photo printer or standard printer via CUPS.
+- Photo preview and capture via multiple camera backends:
+  - **V4L2** - USB Webcams and other Video4Linux2 compatible cameras
+  - **libcamera** - Raspberry Pi Camera Modules (Pi Camera v1, v2, v3, HQ Camera)
+  - **GPhoto2** - DSLR and mirrorless cameras connected via USB
+- **Neural Network background removal** - Remove or replace photo backgrounds in real-time
+- Printout with Canon Selphy photo printer or standard printer via CUPS
 - Configurable image collages with own templates. User can select current template in application.
 - Password protected settings menu:
-	- Copy all photos to USB storage. (currently not supported via flatpak)
-	- Copy collage templates form USB storage. (currently not supported via flatpak)
-	- Disable printing.
-- Brightness of LED preview and flash lights can be adjusted.
+  - Copy all photos to USB storage (currently not supported via Flatpak)
+  - Copy collage templates from USB storage (currently not supported via Flatpak)
+  - Disable printing
+- Brightness of LED preview and flash lights can be adjusted
 
+```
                                  .------------. Current .---------.
                                  | LED Driver |-------->| 20W LED |
                                  '------------'         '---------'
                                     ^
                                     |
                           ENABLE/PWM|          .-----------------------.
                                     |          |                       |
-                                    |          |                       |
-                    .------------------.       |       USB webcam      |
-                    |      Raspi       |<------|        via v4l2       |          -.-.-,~ .   
-                    |    Application   |       |                       |          )     (     
-                    |                  |       '-----------------------'          |_    |     
-                    |     based on     |                                          /(_)---`\   
+                                    |          |  Camera               |
+                    .------------------.       |  - USB Webcam (V4L2)  |
+                    |      Raspi       |<------|  - Pi Camera          |          -.-.-,~ .   
+                    |    Application   |       |    (libcamera)        |          )     (     
+                    |                  |       |  - DSLR (GPhoto2)     |          |_    |     
+                    |     based on     |       '-----------------------'          /(_)---`\   
                     |     QT/Quick     |                                         (_      -'   
                     |                  |           .-------------.                ]      |    
                     |                  |           | Touchscreen |                |    _,')   
@@ -47,88 +52,157 @@ Main features are:
                        |           |
                        `._       _.'
                           "-----"   
-                          
+```
+
 ## Technology
+
 ### Software
-Framework: Qt 6.5 or higher - https://qt.io
+- **Framework:** Qt 6.5 or higher - https://qt.io
+- **Distribution:** Flatpak (primary), native builds supported
+- **Neural Network Runtime:** ONNX Runtime or NCNN for background removal
 
-~~Used camera: Photo camera connected via USB with GPhoto2 - See supported cameras: http://www.gphoto.org/proj/libgphoto2/support.php
-Qt GPhoto2 plugin: https://github.com/dept2/qtmultimedia-gphoto~~ (will be backported later)
+### Camera Support
 
-Flatpak: The main target of this software is the distribution via flatpak. Flatpak can be installed on most of the x86_64 and Aarch64 Linux distributions. It comes with all dependencies needed for execution.
+| Backend | Camera Types | Notes |
+|---------|-------------|-------|
+| V4L2 | USB Webcams, capture cards | Standard Linux video interface |
+| libcamera | Raspberry Pi Camera Modules | Pi Camera v1/v2/v3, HQ Camera |
+| GPhoto2 | DSLR, mirrorless cameras | See [supported cameras](http://www.gphoto.org/proj/libgphoto2/support.php) |
 
 ### Tested Hardware
-Platform: Tested on PC plattform, Raspberry Pi 3B and Raspbarry Pi 4 (Raspberry Pi 3 might also work)
 
-Camera: Tested with ~~Canon EOS 450D and Raspberry Pi camera~~ USB Webca,. Every camera compatible ~~with gPhoto2 ~~or~~ v4l2 will do.
+**Platforms:**
+- PC (x86_64)
+- Raspberry Pi 3B, 4, 5 (aarch64)
+
+**Cameras:**
+- USB Webcams (V4L2)
+- Raspberry Pi Camera Module v2, v3 (libcamera)
+- Canon EOS 450D (GPhoto2)
 
-Printer: Tested Canon Selphy Photo Printer CP910 over wifi via https://github.com/saeugetier/go-selphy-cp. Tested with standard inkjet printer via CUPS.
+**Printers:**
+- Canon Selphy Photo Printer CP910 over WiFi via https://github.com/saeugetier/go-selphy-cp
+- Standard inkjet/laser printers via CUPS
+- Fake printer implementation for testing
+- Printing can also be disabled
 
-Light/Flash: 
+**Light/Flash:**
+LED flash can be driven via Raspberry Pi GPIO. Tested configuration:
 - LED Driver: https://www.aliexpress.com/item/14-37-Inch-LED-LCD-Universal-TV-Backlight-Constant-Current-Board-Driver-Boost-Structure-Step-Up/32834942970.html
 - 20W LED: https://www.aliexpress.com/item/1Pcs-High-Power-10W-20W-30W-50W-100W-COB-Integrated-LED-Lamp-Chip-SMD-Bead-DC/32822371892.html
 
-Display: A touchscreen connected via HDMI is highly recommended.
+**Display:** A touchscreen connected via HDMI is highly recommended.
 
-I2C RTC: If using a Raspberry Pi, it is recommended to use a realtime clock.
+**I2C RTC:** If using a Raspberry Pi, it is recommended to use a realtime clock for timestamping the photos.
 
 ### Housing / Electronics
-My own housing is documented in a seperate git repository: https://github.com/saeugetier/photobooth_hardware
-
-Housing: Plywood 8mm - Cutting via lasercutter. Template generated with http://festi.info/boxes.py/
 
-## Deployment / Installation
+My own housing is documented in a separate git repository: https://github.com/saeugetier/photobooth_hardware
 
-### Flathub ###
+Housing: Plywood 8mm - Cutting via laser cutter. Template generated with http://festi.info/boxes.py/
 
-Install via `flatpak install io.github.saeugetier.photobooth`. Run with `flatpak run io.github.saeugetier.photobooth` or via desktop item.
+## Installation
 
-### Raspbian an a Raspberry Pi
+### Flathub (Recommended)
 
-Minimum Raspbian Buster must be used, which provides QT 6.5 development packages. In order to use GPhoto2 cameras, the Qt GPhoto2 plugin (https://github.com/saeugetier/qtmultimedia-gphoto) must be installed.
+The primary distribution method is via Flatpak. Flatpak can be installed on most x86_64 and aarch64 Linux distributions and includes all dependencies.
 
-For now there is no further support for deployment on Raspbian. Please compile the program yourself from sources.
+```bash
+# Install from Flathub
+flatpak install io.github.saeugetier.photobooth
 
-### Local PC
-
-At least QT 6.5 development packages must be installed in order to compile the application. In order to use GPhoto2 cameras, the Qt GPhoto2 plugin (https://github.com/saeugetier/qtmultimedia-gphoto) must be installed.
+# Run the application
+flatpak run io.github.saeugetier.photobooth
+```
 
-For now there is no further support for deployment on local pc. Please compile the program yourself from sources.
+Or launch via the desktop application menu.
+
+### Raspberry Pi with libcamera
+
+For Raspberry Pi Camera support, ensure your camera is properly configured:
+
+1. Enable the camera in `/boot/firmware/config.txt`:
+   ```
+   # For Pi Camera Module 3
+   dtoverlay=imx708
+   
+   # For Pi Camera Module 2
+   dtoverlay=imx219
+   
+   # For Pi Camera Module 1
+   dtoverlay=ov5647
+   ```
+
+2. Reboot and verify the camera is detected:
+   ```bash
+   libcamera-hello --list-cameras
+   ```
+
+3. Install and run the Flatpak as described above.
+
+### Building from Source
+
+#### Requirements
+- Qt 6.5 or higher development packages
+- OpenCV
+- For GPhoto2 support: libgphoto2-dev
+- For libcamera support: libcamera-dev
+- For neural network: ONNX Runtime, NCNN
+
+#### Build Steps
+
+```bash
+git clone https://github.com/saeugetier/photobooth.git
+cd photobooth
+mkdir build && cd build
+cmake ..
+make -j$(nproc)
+./qtbooth
+```
 
-### Yocto Linux on a Raspberry Pi
+## Configuration
 
-**Deprecated**
+### Configuration File
 
-In order to get the best performance and integration, the recipes for Yocto can be used: https://github.com/saeugetier/poky-photobooth
- 
-Prebuild images will be available soon. For more information how to create an own image for deployment on SD card, please look into the repository.
+The local configuration file is stored in:
+```
+~/.config/saeugetier/qtbooth.conf
+```
 
-## Configuration
+The file contains all application settings including:
+- Selected camera backend and device
+- Neural network runtime settings
+- Printer configuration
+- PIN code for settings menu (default: `0815`)
 
-### Configuration file
-The local configuration file is stored in /home/<user>/.config/saeugetier/qtbooth.conf
-The file contains all application settings and the pin code for the settings menu password protection (only numbers are supported). Default pin code is: 0815
+### Template Files
 
-### Template files
-The local template files for your collage images are stored in /home/user/.local/share/saeugetier/qtbooth
+The local template files for your collage images are stored in:
+```
+~/.local/share/saeugetier/qtbooth/
+```
 
 It contains:
 - Background images for the image collages
-- The "Collages.xml" describing all image collages
+- The `Collages.xml` describing all image collages
 - Border images
 
-The templates can be imported from USB storage. All files with extension "xml,jpg,png,svg" in the folder "layout" will be copied to local template folder.
+Templates can be imported from USB storage. All files with extensions `xml`, `jpg`, `png`, `svg` in the folder `layout` will be copied to the local template folder.
+
+### How to Create Own Templates
 
-### How to create own templates
-Create your own "Collages.xml" file. You can use the file "XmlData.xml" to customize.
+Create your own `Collages.xml` file. You can use the existing file as a reference.
 
-The root node of the XML is named "catalog". It contains nodes for the templates for collages named "collage". The collage must contain at least one "image", a "name", a "background", a "foreground" and an "icon". You can use builtin backgrounds like "WhiteBackground.png" or create your custom one. Forground will be painted in the front layer. So it is highly recommended to have an alpha channel and some cutouts for your photos.
+The root node of the XML is named `catalog`. It contains nodes for the templates for collages named `collage`. Each collage must contain at least one `image`, a `name`, a `background`, a `foreground` and an `icon`.
 
-Images will need information about the position and size. The range of the values for position and size is between 0.0 and 1.0. It is possible to define a image boarder for each image.
+You can use built-in backgrounds like `WhiteBackground.png` or create your custom one. The foreground will be painted in the front layer, so it is highly recommended to have an alpha channel and some cutouts for your photos.
 
-There are optional properties like "printable", which generates a none printable collage (just a single image). So you can use the photobox as an simple camera without printing capabilities.
+Images require position and size information. The range of values for position and size is between `0.0` and `1.0`. It is possible to define an image border for each image.
 
-Example for "Collages.xml":
+Optional properties:
+- `printable` - Set to `false` to create a non-printable collage (single image mode)
+
+#### Example `Collages.xml`:
 
 ```xml
 <?xml version="1.0" encoding="UTF-8"?>
@@ -140,7 +214,10 @@ Example for "Collages.xml":
         <background>WhiteBackground.png</background>
         <foreground>ExampleForeground.png</foreground>
         <images>
-            <image><position x="0.0" y="0.0"/><size width="1.0" height="1.0"/></image>
+            <image>
+                <position x="0.0" y="0.0"/>
+                <size width="1.0" height="1.0"/>
+            </image>
         </images>
     </collage>
     <collage>
@@ -149,25 +226,54 @@ Example for "Collages.xml":
         <background>StarsBackground.jpg</background>
         <images>
             <image>
-                <position x="0.0" y="0.0"/><size width="0.5" height="0.5"/>
-                <border><file>RedBorder.png</file><margin top="10" left="10" right="10" bottom="10"/></border>
+                <position x="0.0" y="0.0"/>
+                <size width="0.5" height="0.5"/>
+                <border>
+                    <file>RedBorder.png</file>
+                    <margin top="10" left="10" right="10" bottom="10"/>
+                </border>
             </image>
             <image>
-                <position x="0.5" y="0.0"/><size width="0.5" height="0.5"/>
-                <border><file>RedBorder.png</file><margin top="30" left="30" right="30" bottom="30"/></border>
+                <position x="0.5" y="0.0"/>
+                <size width="0.5" height="0.5"/>
+                <border>
+                    <file>RedBorder.png</file>
+                    <margin top="30" left="30" right="30" bottom="30"/>
+                </border>
             </image>
             <image>
-                <position x="0.0" y="0.5"/><size width="0.5" height="0.5"/>
-                <border><file>RedBorder.png</file><margin top="20" left="20" right="20" bottom="20"/></border>
+                <position x="0.0" y="0.5"/>
+                <size width="0.5" height="0.5"/>
+                <border>
+                    <file>RedBorder.png</file>
+                    <margin top="20" left="20" right="20" bottom="20"/>
+                </border>
             </image>
             <image>
-                <position x="0.5" y="0.5"/><size width="0.5" height="0.5"/>
-                <border><file>RedBorder.png</file><margin top="50" left="50" right="50" bottom="50"/></border>
+                <position x="0.5" y="0.5"/>
+                <size width="0.5" height="0.5"/>
+                <border>
+                    <file>RedBorder.png</file>
+                    <margin top="50" left="50" right="50" bottom="50"/>
+                </border>
             </image>
         </images>
     </collage>
 </catalog>
 ```
 
-## Issue reporting
-Please use the [issue tracker](https://github.com/saeugetier/photobooth/issues) for bug reporting and feature request. If there are specific bugs or feature request belonging to the Yocto image,  please use [the issue tracker for poky-photobooth](https://github.com/saeugetier/poky-photobooth).
+## Neural Network Background Removal
+
+The application supports real-time background removal using neural networks. Available runtimes:
+- ONNX: currently only CPU supported. Runs best on X86 CPU with more than 4 threads.
+- NCNN: runs best on ARM platform (linited to 4 threads)
+
+Configure the neural network runtime in the settings menu.
+
+## Issue Reporting
+
+Please use the [issue tracker](https://github.com/saeugetier/photobooth/issues) for bug reporting and feature requests.
+
+## License
+
+See [LICENSE](LICENSE) for details.
