Computational photography or computational imaging refers to digital image capture and processing techniques that use digital computation instead of optical processes.
You’re looking at two versions of a photograph taken with an iPhone 7 Plus using the Portrait mode of the built-in camera app. The iPhone 7 camera app creates two versions of the same image, one without the effect, and one with. The one with is obviously the one on the right. These are straight out of the phone, with no other post processing. The iPhone photo was taken using the second 56mm equivalent lens and sensor that comes with the iPhone 7 Plus. The bokeh background blur in the second image was achieved with the iPhone 7 Plus’ computational photographic capabilities.
This bottom photo was taken with my Olympus OM-D E-M5 and the M.Zuiko 12-40mm PRO zoom lens at 32 mm, giving an equivalent focal length of 64mm, at an aperture of f/3.2. Not exact, but close enough.
My biggest take away from this comparison is how well the iPhone Portrait effect computes bokeh compared to my Olympus. If I stick my nose in close enough and really pixel-peep I’m sure I could find issues with the iPhone’s Portrait effect. But then, I’d be ignoring the real point of this, and that’s how a Version 1.0 release of this software is creating an effect that up to this point was achievable only with an optical system. Whether it’s good or not is irrelevant at this point. That fact that a smart phone can achieve an out-of-focus effect very equivalent, if not identical, to what my Olympus can achieve is astounding to me. And here’s the important point to remember; the iPhone’s effect will only get better over time as new releases of the software, with better algorithms, are released. With the Olympus system, I’m “stuck” with what the lens can deliver (which, frankly, I’m quite happy with). And what the iPhone is currently achieving bodes both good and bad for standard photography cameras. This will only ramp up the competition by smart phones against cameras, especially the fixed lens cameras. Perhaps it’s time for general camera makers to wake up and add an equivalent capability to their existing camera lines, especially when using the lower-cost kit zooms.
And while we’re on the subject of computational photography, let’s also mention that micro four thirds has used a very limited version of this since its inception, primarily in distortion (pincushion and barrel) correction. It’s the iPhone which has stepped up the game a few notches with this Portrait effect.
Does this mean the iPhone can replace a camera such as the OM-D? That depends on a case-by-case analysis. It can certainly complement the use of such a camera in overall terms. In an environment with reasonable lighting there’s no reason why you can’t reach for an iPhone 7 Plus as much as you might the E-M5. But when it comes to shooting fast action, or needing low-light capabilities, or faster focusing (and the E-M5 with the 12-40mm is quite fast), or a more fluid shooting situation, then I’d probably reach for my E-M5, assuming it’s within reach at the time.
My iPhone 7 Plus has reached a level of quality and sophistication that I can consider it my backup camera to my Olympus. When I think of a total camera package that allows for immediate post processing as well as immediate sharing, the iPhone 7 Plus with all its integrated capabilities is a better solution than my E-M5. While I do have an SDHC adapter that allows me to read my E-M5 files directly onto my iPhone and iPad, that’s an extra step that can get tiresome, especially if a lot of images are involved. And that adapter is just one more item to loose track of.
I like what I can achieve with the iPhone 7, just like I like what I can achieve with the Olympus E-M5 and the 12-40mm zoom. The iPhone 7 Plus has greatly expanded my technical choices. Now if only my artistic capabilities would grow to match.