If your samples are fixed, you can take a z-stack spanning the entire area you want to capture and then use max intensity projection to collapse them all into one clear image.
But yeah this new camera would be good for living microbes.
The paper describes a split Alvarez (Lohmann) lens [1,2] with a phase modulator between them. I didn't do the math, but it looks like the phase modulator is optically equivalent to a mechanical shift of the Alvarez lenses over regions of the field of view. Alvarez lenses have higher aberrations, and are relatively bulky, compared to normal lenses. AR was referenced in the paper, but this lens will be hard to make compact, and have great image quality, over large fields of view.
It is a new neat idea to selectively adjust focus distance for different regions of the scene!
- processing: while there is no post processing, it needs scene depth information which requires pre computation, segmentation and depth estimation. Not a one-shot technique and quality depends on computational depth estimates being good
- no free lunch. The optical setup needs to trade in some light for this cool effect to work. Apart from the limitations of the prototype, how much loss is expected in theory?
How does this compare to a regular camera setup with lower aperture? F/36 seems excessive for comparison.
- resolution - what resolutions have been achieved? (maybe not the 12 MPixels of the sensor? For practical or theoretical reasons? ) What depth range can the prototype capture? "photo of Paris Arc de triumphe displayed on a screen". This is suspiciously omitted
- how does the bokeh look like when out of focus? At the edge of an object?
The introduction of weird or unnatural artifacts would seriously limit the acceptance
Don't get me wrong - nice technique! But to my liking the paper is omitting fundamental properties
How is this different from using a small aperture size?
When you reduce aperture size the depth of field increases. So for example when you use f/16 pretty much everything from a few feet to infinity is in focus.
Ricoh has some nice EDoF cameras that would make for great QR scanners. They do it the classical way, though. That is, distance dependent chromatic aberration.
In the industrial context you could use a telecentric lens, but it necessarily needs to have large input aperture to have a reasonable field of view.
> Autonomous vehicles might see their surroundings with unprecedented clarity.
This is a pretty good point, which gets me to wonder whether the developers of autonomous vehicles use variable focus adjustments as a part of their ML stack? Or simply set the focal point to infinity.
The image is a byproduct in autonomous driving. Successful implementations (Waymo) use lidar, which doesn’t need focal adjustments. If for some reason quality RGB pixels are needed (e.g. for entity recognition) then they will probably focus on moving objects. This paper ties in nicely with lidar since it apparently needs depth information to work which is exactly what lidar provides
15 comments
[ 3.4 ms ] story [ 44.5 ms ] threadBut yeah this new camera would be good for living microbes.
https://imaging.cs.cmu.edu/svaf/static/pdfs/Spatially_Varyin...
1. https://www.laserfocusworld.com/optics/article/16555776/alva... 2. https://pdfs.semanticscholar.org/55af/9b325ba16fa471e55b2e49...
That would make it really useful, maybe replacing carmera+lidar.
- processing: while there is no post processing, it needs scene depth information which requires pre computation, segmentation and depth estimation. Not a one-shot technique and quality depends on computational depth estimates being good
- no free lunch. The optical setup needs to trade in some light for this cool effect to work. Apart from the limitations of the prototype, how much loss is expected in theory? How does this compare to a regular camera setup with lower aperture? F/36 seems excessive for comparison.
- resolution - what resolutions have been achieved? (maybe not the 12 MPixels of the sensor? For practical or theoretical reasons? ) What depth range can the prototype capture? "photo of Paris Arc de triumphe displayed on a screen". This is suspiciously omitted
- how does the bokeh look like when out of focus? At the edge of an object? The introduction of weird or unnatural artifacts would seriously limit the acceptance
Don't get me wrong - nice technique! But to my liking the paper is omitting fundamental properties
When you reduce aperture size the depth of field increases. So for example when you use f/16 pretty much everything from a few feet to infinity is in focus.
This is a pretty good point, which gets me to wonder whether the developers of autonomous vehicles use variable focus adjustments as a part of their ML stack? Or simply set the focal point to infinity.