Samsung SVP & Head of Automotive Sensors Haechang Lee presented the company's small pixel history and roadmap at SEMI Europe Summit on Sept. 1, 2021. The 576MP mobile sensor is planned for 2025:
Is there a strong business case for such high resolutions?
The reason I ask, is that while pixel binning and computation can be used to great effect, the resolution is often downsized for the location of where most images end up. Lets say 60% end up on social media, 5% might be printed, 5% are printed by people looking for professional quality and 30% never see the light of day and sit on a hard drive for years.
For that end use case (someone will have real-world stats), is 575MP warranted given the downsizing for Facebook etc? Likewise, last time I got something printed it needed to be 300 or 600dpi for the size of paper, meaning all those extra pixels were down-sampled anyway.
I agree there are large advantages to higher spatial sampling and some interesting tricks we can do in post. I also agree people might zoom in etc, but to this degree? I'm sure a 50MP image will be more than enough for 99.9% of end use cases.
As an example, the other day I used my phone to take a picture of my electricity meter. I also used it to take a picture of some pine bookshelves for sale in the classified adds. Neither warrant anything above 10MP.
Previous commentators should bear in mind that this enormous image resolution does not serve conventional purposes, but is designed solely for virtual reality applications in the upcoming 6G wireless internet era. The other day, for example, Samsung presented a scenario in which 6G ecosystem users use a holographic clone of themselves to interact with others. These extraordinarily high image data resolutions are designed with such future applications in mind.
More pixels doesn't change the nature of visible light. Lenses are still diffraction limited. The extra pixels add a lot of processing, transport, and storage burden without adding much, if any information.
Can you show me the MTF of the lens or better the objective that will go in front of this sensor? I bet that it is not capable to resolve this pitch. And the MTF of the system is not only the sensor. So yeah future applications will require in any case that the objective is capable to keep up with the sensor otherwise is just marketing, bigger is better no?
On December 1st 2020 Samsung Research published their white paper, which includes scenario illustrations in which 6G ecosystem users access "digital twins" of themselves in interaction with others.
This insightful research paper can be found at https://cdn.codeground.org/nsr/downloads/researchareas/20201201_6G_Vision_web.pdf
They outline three key 6G services: Immersive extended reality (XR); high-fidelity mobile hologram; and digital replicas.
Report Quotes:
"It will be possible to replicate people, devices, objects, systems, and even places in a virtual world."
"With the help of advanced sensors, AI, and communication technologies, it will be possible to replicate physical entities, including people, devices, objects, systems, and even places, in a virtual world."
"In a 6G environment, through digital twins, users will be able to explore and monitor the reality in a virtual world, without temporal or spatial constraints. Users will be able to observe changes or detect problems remotely through the representation offered by digital twins."
Is there a strong business case for such high resolutions?
ReplyDeleteThe reason I ask, is that while pixel binning and computation can be used to great effect, the resolution is often downsized for the location of where most images end up. Lets say 60% end up on social media, 5% might be printed, 5% are printed by people looking for professional quality and 30% never see the light of day and sit on a hard drive for years.
For that end use case (someone will have real-world stats), is 575MP warranted given the downsizing for Facebook etc? Likewise, last time I got something printed it needed to be 300 or 600dpi for the size of paper, meaning all those extra pixels were down-sampled anyway.
I agree there are large advantages to higher spatial sampling and some interesting tricks we can do in post. I also agree people might zoom in etc, but to this degree? I'm sure a 50MP image will be more than enough for 99.9% of end use cases.
As an example, the other day I used my phone to take a picture of my electricity meter. I also used it to take a picture of some pine bookshelves for sale in the classified adds. Neither warrant anything above 10MP.
Binning is cheating.
ReplyDeleteDo you need 1GB internet? To check email, let's just do dial-up 55.6kbps.
ReplyDeleteMarketeers and their business cases. They should go back at selling bananas to monkeys
ReplyDeleteAnd then Apple makes it first š¤£
ReplyDeletePrevious commentators should bear in mind that this enormous image resolution does not serve conventional purposes, but is designed solely for virtual reality applications in the upcoming 6G wireless internet era.
ReplyDeleteThe other day, for example, Samsung presented a scenario in which 6G ecosystem users use a holographic clone of themselves to interact with others.
These extraordinarily high image data resolutions are designed with such future applications in mind.
More pixels doesn't change the nature of visible light. Lenses are still diffraction limited. The extra pixels add a lot of processing, transport, and storage burden without adding much, if any information.
DeleteLink?
DeleteCan you show me the MTF of the lens or better the objective that will go in front of this sensor? I bet that it is not capable to resolve this pitch. And the MTF of the system is not only the sensor. So yeah future applications will require in any case that the objective is capable to keep up with the sensor otherwise is just marketing, bigger is better no?
Deletehttps://www.technocracy.news/samsung-says-6g-will-bring-digital-twins/
DeleteI must correct myself:
DeleteOn December 1st 2020 Samsung Research published their white paper, which includes scenario illustrations in which 6G ecosystem users access "digital twins" of themselves in interaction with others.
This insightful research paper can be found at https://cdn.codeground.org/nsr/downloads/researchareas/20201201_6G_Vision_web.pdf
They outline three key 6G services:
Immersive extended reality (XR); high-fidelity mobile hologram; and digital replicas.
Report Quotes:
"It will be possible to replicate people, devices, objects, systems, and even places in a virtual world."
"With the help of advanced sensors, AI, and communication technologies, it will be possible to replicate physical entities, including people, devices, objects, systems, and even places, in a virtual world."
"In a 6G environment, through digital twins, users will be able to explore and monitor the reality in a virtual world, without temporal or spatial constraints. Users will be able to observe changes or detect problems remotely through the representation offered by digital twins."