Unlike most phones released these days, Apple's new iPhone SE has a 12 megapixel rear camera lens. However, one of the cameras found by fellow Dieter Bohn in his iPhone SE review turned out to be still pretty good in bright lighting conditions.
For portrait photography, the iPhone SE's camera uses machine learning to estimate the depth of field, and one of the developers of the mobile photography app Halide, Ben Sandofsky, took a closer look at how portraits taken with SE's single camera actually work.
The main feature of iPhone SE's portrait mode is that it can perform a feature called “single depth estimation” that is activated by the iPhone SE's A13 Bionic processor (same processor on iPhone 11 and 11 Pro). This processor and depth estimation allow you to use a phone capture depth map for photos, unlike the iPhone XR with a single lens. Sandofsky found that SE may also estimate the depth of flat pictures.
Sandofsky, for example, took this picture with the iPhone XR and iPhone SE.
Image: Halide
In this image slider you can see how to measure the depth of both phones. The iPhone XR's depth map is on the left and the iPhone SE is on the right.
The depth map in iPhone SE doesn't perfectly represent the depth of the real scene in any way, but it's great that machine learning can be used to estimate the SE's real depth in flat pictures.
I highly recommend taking a few minutes to read Sandofsky's blog as a whole. He has more examples of real depth mapping, including a more detailed explanation of how SE's camera captures depth, and finding depth in a 50-year-old grandmother's slide film!
If you want to see more of his thoughts on SE's camera, check out Dieter's review.