Blog | Extracting Data from Apple Photos: The Power of Organization

Extracting Data from Apple Photos: The Power of Organization

I have a massive archive of photos from the last 20+ years of shooting. I have hundreds of shoots and hundreds of thousands of images that I have never published because it’s so time-consuming to sort through them, choose the best shots, caption, and tag them.

I decided to put an end to the manual work that comes after the fun of photography.

I’ve spent my limited spare time over the last month or so building a new application that massively reduces the friction involved in sharing photos. This app takes the hassle out of photo management by utilizing the power of the Apple Photos app as the source of truth.

A few months ago, I imported my entire archive in the Photos app. I converted all the RAW files (preserving the originals, of course) to JPEGs before importing to save on file size.

Apple Photos does some fascinating stuff behind the scenes with your photos once you upload them using on device AI/ML. All the raw data is stored in a local SQLite database which you can access using an excellent python tool and library called osxphotos from Rhet Turnbull

My app uses osxphotos to pull information from your Apple Photos Library, including data such as face detection, tags, detected text, location data, albums and even Apple’s own AL/ML Photo scores. This data is then used to create this website you’re looking at right now.

How it Works

The Apple Photos app acts as the central hub for all of your photo information. Thanks to advanced machine learning algorithms, the app can accurately detect faces in your photos, making it easy to search for specific people. It also automatically assigns labels to your photos, based on the content of the image. This means that you no longer have to spend hours manually captioning and organizing your photos.

In addition to tags and labels, the app also uses text detection to identify the subjects of your photos. This makes it even easier to search for specific photos, based on the subjects that are most important to you. For example, if you have a photo of a beautiful sunset, the app will automatically detect the keyword "sunset" and tag the photo accordingly.

Another great feature of the app is the ability to reverse geocode the location of your photos. This means that the app can take the GPS coordinates of a photo and turn them into a readable address. This is especially useful for travel photos, as it allows you to easily see where each photo was taken.

The app also allows you to organize your photos into albums, making it easy to find specific photos based on events or themes. You can also leave comments on your photos, mark your favorite photos, and even assign scores to your photos, based on their overall quality.

Once I have the Apple Photos data extracted, I enhance it using a couple of popular tools from AWS and OpenAI to find celebrities and write (often hilariously wrong) captions.

Share Your Photos with the World

The best part of this app is the ability to share your photos with the world on your domain. This means that you can have a personalized photography website, complete with all of your best photos, without having to deal with the hassle of manual organization. Nor do you have to upload your work to a platform that’s whole purpose is to make money from your content via advertising .

In conclusion, the new app I built makes photo management and sharing a breeze. By utilizing the power of the Apple Photos app, you can easily access all of your best photos in one place, without having to spend hours captioning and organizing them.

Over the coming days, I’ll write more about the specific parts of the app and the technologies I’ve used to create it. Keep in eye out for that here on the blog.

Beta Testers

I’m looking for some folks who are interested in trying out the app. Reach out if you’re interested.


  • Apple Photos
  • Photography
  • Photo Management
  • Photo Sharing
  • osxphotos
  • Face Detection
  • Tags/Labeling
  • ML Keyword Detection



Post date:

Tuesday, February 14th, 2023 at 4:14th:02 PM