ImageViewer

Cross Platform Image Viewer

An open source image viewer written in JavaFX. Create local image collections for recursively looking through directories on your local system. Images are indexed and sorted based on jpeg exif date. It's probably best to add directories with shallow depth. Currently it will recursively search with a hardcoded depth of 10. The original use case was to explore images on a hard disk that were in the following directory structure.

• root
  • 2011
    • 2011-03-28
  • 2020
    • 2020-shoot1
      • 2020-06-13
      • 2020-06-14
    • 2020-shoot2
      • 2020-07-01
      • 2020-07-02
  • 2021
    • 2021-09-16

An example screenshot of the application displaying the Flickr2k dataset

Performance

When adding new directories, especially directories with many images, indexing will occur on ALL images and may use a significant amount of CPU. Once indexing is completed subsequent loads of the application are fast.

The app is a memory hog even if -Xmx500m is supplied for max heap space. Memory usage could likely be improved by reducing in memory image cache sizes, however the cache needs to be large enough to hold all on screen images, which can be up to 2,000 (64x64) images when zoomed out.

File System

Source images will never be modified or deleted. Source images will only be read. This is an image viewer, not an image editor.

Image indexing cache and database files are located in the following directory (platform dependent).

• macos: {USER_HOME}/Library/Application Support/com.onebyte_llc.imageviewer/
• windows: {USER_HOME}\AppData\Local\com.onebyte_llc.imageviewer\
• linux: {USER_HOME}/.com.onebyte_llc.imageviewer/

Image Formats

Currently only jpeg and png images are supported. To add other image formats a ImageLoader and TypeDefinition need to be added in com.onebyte_llc.imageviewer.backend.image. Then the new ImageLoader needs to be registered in the backend Context constructor. An example of this is JpegImageLoader and JpegImageTypeDefinition.

Slideshows

A slideshow can be started and stopped by clicking on an image and then pressing the space key.

Running

To run the java project on any operating system

./gradlew run

Building an OS app image

To build a native app image run the following on the target OS. It should work on macOS, Windows, and Linux. It will create a standalone executable.

./gradlew jpackageImage

Future Plans

• Support video formats (probably using VLC)
• Support User configurations, examples,
  • allow user to set depth of source image search on a folder
  • allow user to set duration an image shows during a slideshow
  • etc.
• Support Filtering/sorting images by
  • image creation date
  • image modified date
  • etc.
• Support Face detection
  • this will allow users to name faces and then search/filter

License

This software is licensed under GPLv2