What it is and the problem it solves
Computational photography is digital image capture and processing that substitutes computation for optical processes. It solves the problem of physical optical limits—like numerical aperture—and enables capabilities impossible with film or conventional optics.
How it works
It replaces optical processes with digital computation during image capture and processing. It uses coded aperture, coded exposure, and light field imaging — each modifying physical capture to make inverse problems like deblurring or de-focusing mathematically tractable.
What works
Light field imaging delivers post-focus and enhanced depth-of-field. Coded exposure makes motion deblurring well-conditioned. Lens-based coded aperture with broadband masks makes out-of-focus deblurring well-conditioned. Coded aperture improves image quality in X-ray and astronomy.
What does not
It does not eliminate the need for optical elements in all cases—only some implementations obliterate them. It does not deliver post-focus or enhanced depth-of-field without novel optical elements in light field systems. It does not guarantee improved image quality outside its verified domains: astronomy, X-ray, THz, or motion-deblurred photography.
What it changes
It changes camera design by decoupling function from lens physics—enabling smaller, cheaper, or more capable systems where conventional optics fail. It shifts focus from hardware correction to algorithmic reconstruction.
Is it worth your time
Yes—if you design imaging systems, work in X-ray or THz domains, or build cameras where size, cost, or mechanical focusing are constraints. No—if you rely on off-the-shelf optics and do not control the full stack from sensor to reconstruction.
