BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Australia/Brisbane
X-LIC-LOCATION:Australia/Brisbane
BEGIN:STANDARD
DTSTART:19920301T030000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
TZNAME:AEST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20191203T080509Z
LOCATION:Plaza Meeting Room P1
DTSTART;TZID=Australia/Brisbane:20191120T112100
DTEND;TZID=Australia/Brisbane:20191120T114200
UID:siggraphasia_SIGGRAPH Asia 2019_sess118_papers_190@linklings.com
SUMMARY:Reparameterizing Discontinuous Integrands for Differentiable Rende
ring
DESCRIPTION:Technical Papers\n\nReparameterizing Discontinuous Integrands
for Differentiable Rendering\n\nLoubet, Holzschuch, Jakob\n\nDifferentiabl
e rendering has recently opened the door to a number of challenging invers
e problems involving photorealistic images, such as computational material
design and scattering-aware reconstruction of geometry and materials from
photographs. Differentiable rendering algorithms strive to estimate parti
al derivatives of pixels in a rendered image with respect to scene paramet
ers, which is difficult because visibility changes are inherently non-diff
erentiable.\n\nWe propose a new technique for differentiating path-traced
images with respect to scene parameters that affect visibility, including
the position of cameras, light sources, and vertices in triangle meshes. O
ur algorithm computes the gradients of illumination integrals by applying
changes of variables that remove or strongly reduce the dependence of the
position of discontinuities on differentiable scene parameters. The underl
ying parameterization is created on the fly for each integral and enables
accurate gradient estimates using standard Monte Carlo sampling in conjunc
tion with automatic differentiation. Importantly, our approach does not re
ly on sampling silhouette edges, which has been a bottleneck in previous w
ork and tends to produce high-variance gradients when important edges are
found with insufficient probability in scenes with complex visibility and
high-resolution geometry. We show that our method only requires a few samp
les to produce gradients with low bias and variance for challenging cases
such as glossy reflections and shadows. Finally, we use our differentiable
path tracer to reconstruct the 3D geometry and materials of several real-
world objects from a set of reference photographs.\n\nRegistration Categor
y: Platinum Pass, Full Conference Pass, Full Conference One-Day Pass
URL:https://sa2019.conference-program.com/presentation?id=papers_190&sess=
sess118
END:VEVENT
END:VCALENDAR