Panasonic: Computation Photography not needed

Bobapingu wrote:

This is a case of Chinese whispers. I've read what Pana has said. My interpretation of that is that Pana is saying computational photography is not a priority for them at the moment. That's it, nothing more.

PetaPixel has interpreted it as follows, "The company has made its stance on broader implementation of computational photography known, but the door isn’t closed on all features that technically fall under that banner."
That's a little stronger and a bit inflammatory in saying, "The company has made its stance..."

The OP has interpreted the PetaPixel comment as, "Panasonic: Computational Photography not needed",
Getting stronger.

Now we have posters making the jump to, "Panasonic says no to computational photography".

We need to cool off a bit.

OMS is the leader in computational photography.

It started well before m43 even existed. For example in medical imaging or astronomy if you think of things like image stacking. I give you just one example: for cancer treatment of say a brain tumor with a linear accellerator, you need precise control of the beam shape and direction. Well before m43 existed, CT, Xray and MRI images, each containing different information, and some with extremely distorted geometry, needed to be superimposed and stacked into one single image of very precise dimensions to scale, to be useful for guidance of the beam. The same was needed for 3d guided surgery, because as a surgeon you cannot differentiate visually a brain tumor from healthy brain tissue. The surgeon's knife is fitted with two reflective balls which are tracked by two cameras, and on a screen the surgeon can see his knife superimposed to the stacked CT/MRI/Xray images to see where he needs to cut. And that was in the late 90's when I worked with such equipment. Back then this was hi-tech, even it was only the beginning of computational imaging.

Phone cameras is where computational photography for consumer goods started. Many years ago. Think image stacking (focus stacking, or portrait background blur)

Yes Olympus was first at implementing digital ND filters in a consumer camera (the EM1X which had two image processors inside to provide the computational power). Following the ideas published by the German Fraunhofer Institute several years earlier. Today it's in the Samsung S23 Ultra phone too.

At the moment, there doesn't seem to be a great rush to follow. It'll probably come but why should Pana be the first to chase OMS?

Panasonic makes smartphones with cameras too, since 2004. One would think they have a pretty good understanding of what computational photography can do for their smartphones. Maybe they think that ILC camera users will use a phone, if they want the more extreme sorts of computational photography like background blur and worse. I kinda agree.

Because it's m4/3? What has that got to do with it?

Well, to some extent it actually does. You see, smaller sensors have higher readout speeds, and therefore can have higher frame rates than larger sensors. They also produce smaller raw files, that can be processed faster by a given CPU power.

Both these effects make it easier and cheaper to do things like focus stacking, HR, HHHR, digital ND filters and the like in the smallest of all ILC formats, which happens to be m43. Hence it should be no surprise those features first appeared in m43, and not in larger formats. But even in m43, the processing power needed for these features required an extremely huge and heavy camera to be purpose built just for this, with two separate image processors inside, and a heat pipe for cooling - it was the EM1X of 2019. A commercial flop as a camera, but an important technology carrier for future developments such as the EM1.3 of 2020.

Why Olympus and not Panasonic?

Remember, 2018 was the time that Panasonic joined the L-mount alliance. They were for a few years very busy creating L-mount cameras and lenses, and that put m43 R&D on the backburner.... actually until today. And it turned out to be a very wise decision, thanks to L-mount Panasonic is still the number 5 in camera market share. Without L-mount, they would today be completely insignificant.

And another important reason is video: Panasonic puts way more of it's efforts into m43 hybrid cameras, rather than still cameras. And most of these new computational features simply do not work at all for video. Hence they are commercially uninteresting for Panasonic - they won't sell more cameras with new (and somewhat controversial) features that only work for stills, when the vast majority of their customers want to make videos.