undefined | Better HN

0 pointsmirsadm3y ago0 comments

It does not take hundreds of exposures. That is impractical and above all not a good idea. In low light the shutter speed is likely slow (1/10th of a second for a example). Even 16 frames is more than a seconds worth of frames. At that point the differences start to become too high for stacking to be worthwhile, especially because it is hand held. Secondly, the reduction of noise by merging is at best the sqrt (number of frames). Past even 12 frames the benefits start to be questionable.

0 comments

Xcelerate3y ago

> Secondly, the reduction of noise by merging is at best the sqrt (number of frames)

This assumes the (typically Gaussian) noise is applied to a static image. Arguably, one could exploit the slight shakiness in a handheld shot to create an image with even less noise.

As a thought experiment, consider thousands of shots of a perfectly static scene made with an idealized, noiseless camera that is moved a very tiny amount for each shot. You could continue improving the resolution of a generated, composite image quite a bit until warping due to camera displacement became noticeable.

Recent techniques like this are actually being used for cryo-electron microscopy to create extremely high-resolution imaging of proteins.

mirsadmOP3y ago

Yes what you are referring to is implemented already by Google and probably others. The Google camera uses it to skip demosaicking by shifting the image by a pixel using OIS. Still the original post was under the impression that hundreds of exposures are merged which is just not the case

j / k navigate · click thread line to collapse