One of the most important reasons that made cameras more  user-friendly was the ability to focus automatically without requiring  the photographer to tilt the focus ring back and forth to get sharp  image. The “magic” trick was the contrast within the frame. Traditional  contrast-based autofocus systems simply had to scan the entire focus  range (i.e. from the minimum focus distance to infinity) and luck the  focus on the area with the highest level of contrast. This process was  relatively slow, taking anywhere between 500 ~ 2000 milliseconds.  Clearly, this left much room for improvement. Another problem was  focusing on moving objects. The slow autofocus meant that the shot could  be missed entirely while the camera “hunts” for focus.

Phase Detection Autofocus   

Phase detection is a technology that utilizes special sensor to  detect focus instead of relying completely on the contrast within the  frame. The traditional phase detection in professional DSLRs (i.e.  cameras with mirror to direct light to the optical view finder) allows  some of the light to bypass the mirror (the mirror is about 5%  transparent) to hit the phase detection sensor. The advantage is much  better autofocusing at the expense of some light not being used for the  actual taking of the photo.

Mirrorless cameras can’t have this luxury since they don’t  incorporate mirrors. Instead, special photocells are added to the image  sensor, which are used for autofocusing.

Phase detection has greatly improved autofocusing by “sacrificing” a  small amount of light. The biggest challenge (or disadvantage if you  like) is when focusing in low-light conditions. The amount of available  light for focusing may be inadequate, forcing the camera to fallback to  contrast-based autofocus.

Depth from Defocus

Depth from Defocus, or DfD for short is a technique that relies on  determining the focus area from the out-of-focus data. The Camera takes  multiple out-of-focus pictures and applies complex algorithm to work out  the best focus point. The theory behind DfD emerged in the late 80’s,  however, it required massive amount of computation, making it  prohibitively expensive.

In 2012, Panasonic released the Lumix GH4, which was the first  mirrorless camera to incorporate DfD technology. What is particularly  interesting about Panasonic’s implementation is the inclusion of an AI predictive algorithm that predicts the point of focus based on the distance and velocity of  the object of interest, promising extremely fast and accurate focusing.  Panasonic claims that their DfD can focus within 70 milliseconds only.

Additional advantage is that by eliminating the dedicated focusing  sensors, no light is sacrificed. This is especially more critical for  the small Micro Four Thirds sensor. This also allows the camera to  autofocus in a very dim light down to -4 EV (Exposure Value) because all  the light is used for focusing.

What really made DfD possible was the multi-core, ultra-fast and  power-efficient processing engine. In fact, the engine has to perform a  massive amount of calculations to reliably detect the focus area in a  very short amount of time before falling back to pure contrast-based  autofocusing if DfD failed.

However, Panasonic’s latest Venus engine is still not capable enough  to reliably focus while taking video. With a vast stream of data needed  to capture video at 4K resolution, the engine falls short and the camera  is forced to hunt for focus. Annoyingly, this “hunting” is visible and  can ruin the shot.

By incorporating AI into photography, a new chapter of possibilities  has opened. The main obstacle today is the amount of processing power a  camera can have. However, if Moore’s law holds out, Depth from Defocus  technology is very likely to prevail in all applications, including  video and DSLRs may as well disappear. After all, why carry a large and  heavy camera around when a smaller and light one can do the job, better  and smarter?

Note: This article first appeared on CognitionX blog