Twitterrific Adds Face Detection To Improve Image Crops
It means that as Twitterrific displays media thumbnails in the timeline (pictures, videos, etc), the app tries to detect faces and frame the thumbnail so faces are always showing. In short, if Twitterrific sees a face in a tweet, it tries to make sure you see it too!
The effect when scanning through your list of tweets in the timeline can be dramatic. Previously Twitterrific always framed thumbnails on the center of images, but many times people’s faces aren’t in the middle, especially on portrait shots. Check out these before and after comparison screen shots to see the difference facial framing makes in the timeline:
Apple includes a load of APIs as standard in the SDK, ranging from language analysis, physics engines to image facial feature recognition. As a developer, you are always looking at ways to develop new apps and features by applying these — rather niche but rich — frameworks. Often, this means creating an entirely new app off the back of such a framework (Pedometer++ is a great example) but Twitterrific have cleverly thought up a way of using Core Image detectors to enhance the experience of their existing app.
They use Apple’s face detection APIs to better frame cropped images in the timeline. Is it life changing? No. Is it nice? Yes. Ideally, the feature is invisible. This is why these frameworks exist. The Iconfactory can piggyback off the work done by Apple (which originally developed the face detectors for its own Camera app) and deliver improvements like this to users with relatively little effort than if they had to develop the face detection engines in house.