Web-based photo organiser that has AI auto-tagging and mobile apps. It can be used in conjunction with NextCloud - simplest way is to mount the data folder as a Docker volume for Photonix.
It’s based on web technologies. Run it on your home server and it will let you find what you want from your photo collection using any device. Smart filtering is made possible automatically by object recognition, location awareness, color analysis and other algorithms.
To have an organized collection of photos we believe each one should be tagged with as much useful information as possible. Tags are different from categories; a single photo can have many tags but could only be in a single category. This is why we think tagging is a lot better than trying to decide on a descriptive folder name to store it in. Having a database allows super fast filtering of photos by tags.
There are several different sources for these tags:
- Metadata that the camera produces — aperture, shutter speed, GPS location, date and time are examples of these.
- Features that the user provides because they anticipate searching for them later — this could be things like a family member’s name, a holiday where photos were taken or objects in the photo.
- Smart features that a computer has learnt to identify — these could be objects that were detected, styles that are similar, colors that are visible and so on.
We think our photo management software excels in the area of smart features. In recent years, as you have probably seen in the news, there have been dramatic breakthroughs in the field of artificial intelligence (AI), machine learning and deep learning. It was these developments that inspired me to embark on this project. Google Photos has demonstrated what is possible in the area of object recognition and detection but there is a lot more that can be done. Google has been very open with their technology and we have been able to leverage some of the same frameworks and pretrained models that they have created.
We currently provide the following smart tagging classifiers:
- Object — Based on Google’s models but runs entirely within Photonix so no images are shared externally. The bounding box of each image is also stored, which can be used to determine its significance by size. This in turn helps with ordering results that are returned.
- Style — Trained on an image dataset labelled according to artistic styles such as geometric, minimalist, noir. This is quite a curious tool to play with and might bring some serendipitous results.
- Color — Starting with a simple palette, it finds the nearest color for each part of an image. The area of the each color within an image is used to calculate a significance weighting for each.
- Location — Currently works off GPS data that cameras and phones often produce. A built-in map of the world allows it to determine country and nearest big city to tag it with. We aim to look at more intelligent feature recognition possibilities in future for photos that aren’t GPS tagged.
Quickly view images full-screen on any device you have logged in via app or web browser. Metadata about the photos is easily accessible by just scrolling down. This includes the standard details about the camera settings and date/time information but also extended data that the smart AI has determined. This includes the colors that were identified, objects in the photo, the county name and location on a map and more is yet to come.
Tracking multiple versions of photos and de-duplication
When you get into editing (with Lightroom, Darktable, Photoshop, RawTherapee, PaintShop Pro etc.) you quickly realize how easy it is to get into a mess with duplicate photos and different types of files. Free and paid photo management software is often lacking in this area and will leave you spending hours to come up with a system and then sorting by cutting and pasting. This has led many people to set up folders of differing types and versions but it can often be difficult to then go back and find the RAW version or different edits of a photo later on.
It would be great if some amazing photo organizing software could do that job for you! Well, Photonix doesn’t just blindly show every photo it finds, it compares date/time and other metadata to identify which files are actually versions of another. It will show what it thinks is your latest edit as the preview/thumbnail image but RAW files, camera JPEGs and previous edits are never more than a couple of clicks away in the user interface. You’ll soon be able to select a preferred edit if you prefer a previous version to be displayed.
Android and iPhone apps
We now have Photonix mobile apps available to download in the Google Play Store and Apple App Store. These apps give quick access to your server from your phone’s home screen. More features will be coming soon such as the ability to automatically upload new photos from your camera roll as you take them. You’ll then be reassured that your photos are safe and backed-up to your home server or cloud storage.
Please note that these mobile apps will not provide any useful functionality without having a server to connect to. If you just want to play with the app or download it as a reminder, you can connect it to our demo server. You can help our project succeed by leaving reviews in the app stores.
Maybe you’re setting up a home server running Photonix for your family. Maybe you’re thinking of using Photonix in your business processes with your colleagues. You can each have your own private library and then a separate shared library for collaboration.
Typically there will be one admin user created at the start and this user can then create additional accounts later on. Users can have a variety of roles and restrictions. Users will be able to invite other users to view/collaborate on a photo library. Access to the shared library can be controlled per user, i.e. read-only or read-write.
We use a framework called Django that means we’re not reinventing code to do things like authentication and authorization, improving security. All users will be able to use the full photo management features for searching and, if they have write permissions, the ability to tag and create albums.
We believe sharing should be easy and quick, whether it’s a single photo you want to send to your parents or whole photo albums you have prepared for a client shoot (e.g. a wedding). Unlike software such as Magix Photo Manager and Movavi Photo Manager, Photonix is built to be online.
Being online also comes with its own concerns around security and privacy. Your image files are very important and should be protected at all cost. We have experience building apps which handle sensitive user data and we make this the number one priority with everything we build. This is in contrast to some services that treat it as an afterthought.
We want to do this right and gain your trust, providing transparency about our procedures to security experts, developers and technically-minded journalists. This is one of the reasons that Photonix is open source — we’re so confident in our code that we’re happy for anyone to take a look.
We hope you’ll understand how seriously we take this and will bear with us until the feature is fully ready. Please sign up to the mailing list to be updated when this is ready.
Presentations and slideshows
It’s important to be able to show your images to people and guide them through your story and there are several scenarios where you may want to do this. Say you are at your home or the home of a family member, the living room TV is likely to be the biggest and best screen to use — if only it was convenient. Perhaps you are discussing your photos with a client and want to quickly present to them. These are a couple of experiences that we believe should be better.
The dual-screen scenario inspires a lot of creative ideas for us. Your phone could be controlling a TV via an app whilst still allowing you to search for the next photo. A client could be viewing a photo you just sent them in their web browser while you’re deciding on the next one to show.
It could be a very fluid and engaging experience. It will take a while for us to connect all these different technologies into a seamless experience but hopefully you can see the benefit.
We aim to add limited support for video files soon but right now, we’re trying to do the best job we can for photographers.
The Android app has 9 trackers --- must be a record for an open-source mobile app:-
Trouble is, there's no iOS equivalent of Exodus Privacy (or similar) AFAIK --- and Apple does not verify what the dev professes:-
"The developer, Epix Studios Limited, indicated that the app’s privacy practices may include handling of data as described below. This information has not been verified by Apple."
I've opened an upstream issue for the Android trackers:-
@hillside502 interesting response from the developers!
That link looks concerning. There aren't any trackers that we added. Maybe Expo or React Native tools add stubs by default for these - quite worrying. Would be nice if there were more details. I'm guessing code signature means that it found strings for things like com.facebook.adsin the APK. I'll try and reproduce. Thanks for highlighting.
Looks like they've been using the Expo build service, whose own app has 10 trackers:-