BookNotesSlidesP-sets Syllabusdraft · v0.2.49.1
💬Comments welcome. To leave a note, select any text and click the note / highlight button that pops up — or open the panel with the tab at the top-right (‹). Notes are visible only inside our private review group.
Computational Photography, an AI-powered Slopendium — 16 Deblurring
expand to📖 Full book outline1 parts · 10 chapters · 30 sections · 0 figures embedded · 0 placeholders · double-click a figure to enlarge
▸Part 16 DEBLURRING⧉
<!--uid: deblurring-->
▸16.1 What blur is, and what deblurring can undo⧉
16.1.1 Why naive inversion fails⧉
16.1.2 The fix, stated once⧉
16.1.3 Which space to deblur in⧉
▸16.2 Non-blind deconvolution⧉
16.2.1 The Wiener filter: the regularized, noise-aware inverse⧉
16.2.2 Richardson-Lucy: iterate toward the likelihood⧉
16.2.3 Constraints instead of priors⧉
▸16.3 Priors for deconvolution⧉
16.3.1 Sparse gradients: the prior that keeps edges sharp⧉
16.3.2 Solving it, and the trick that made it fast⧉
▸16.4 Blind deblurring⧉
16.4.1 Estimating the kernel and the image at once⧉
16.4.2 Can you measure the kernel instead?⧉
16.4.3 Why the obvious optimizer fails⧉
16.4.4 Two ways to solve it⧉
▸16.5 Spatially varying blur⧉
16.5.1 Rotational camera shake⧉
16.5.2 A table of blurs, and which are convolutions⧉
16.5.3 What ships in a phone⧉
▸16.6 Defocus, and undoing what the lens softened⧉
16.6.1 The defocus PSF, and why one deconvolution is wrong⧉
16.6.2 Estimating the blur field⧉
16.6.3 Lens softening, which is a different problem with the same shape⧉
16.6.4 Engineering the blur instead of fighting it⧉
▸16.7 Learned deblurring⧉
16.7.1 Networks that skip the kernel⧉
16.7.2 The datasets are the argument⧉
16.7.3 What learning did and did not change⧉
▸16.8 Video, bursts, and lucky imaging⧉
16.8.1 Atmospheric seeing: why one long exposure fails⧉
16.8.2 Shoot thousands, keep the sharpest⧉
16.8.3 Align and stack the survivors⧉
16.8.4 A poor man's adaptive optics⧉
▸16.9 Measuring a deblur⧉
16.9.1 What went wrong, specifically⧉
16.9.2 What to measure⧉
16.9.3 On what data⧉
16.9.4 What to report⧉
16.10 Deblurring TLDR⧉