BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260817T171530Z
LOCATION:Hall B5 (1)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241203T151900
DTEND;TZID=Asia/Tokyo:20241203T153100
UID:siggraphasia_SIGGRAPH Asia 2024_sess106_papers_466@linklings.com
SUMMARY:DiffCSG: Differentiable CSG via Rasterization
DESCRIPTION:Haocheng Yuan (University of Edinburgh); Adrien Bousseau (Inri
 a Sophia-Antipolis, Université Côte d’Azur); Hao Pan (Microsoft Research A
 sia); Quancheng Zhang (Nanjing University); Niloy J. Mitra (University Col
 lege London (UCL), Adobe Research); and Changjian Li (University of Edinbu
 rgh)\n\nDifferentiable rendering is a key ingredient for inverse rendering
  and machine learning, as it allows to optimize scene parameters (shape, m
 aterials, lighting) to best fit target images. Differentiable rendering re
 quires that each scene parameter relates to pixel values through different
 iable operations. While 3D mesh rendering algorithms have been implemented
  in a differentiable way, these algorithms do not directly extend to Const
 ructive-Solid-Geometry (CSG), a popular parametric representation of shape
 s, because the underlying boolean operations are typically performed with 
 complex black-box mesh-processing libraries. We present an algorithm, Diff
 CSG, to render CSG models in a differentiable manner. Our algorithm builds
  upon CSG rasterization, which displays the result of boolean operations b
 etween primitives without explicitly computing the resulting mesh and, as 
 such, bypasses black-box mesh processing. We describe how to implement CSG
  rasterization within a differentiable rendering pipeline, taking special 
 care to apply antialiasing along primitive intersections to obtain gradien
 ts in such critical areas. Our algorithm is simple and fast, can be easily
  incorporated into modern machine learning setups, and enables a range of 
 applications for computer-aided design, including direct and image-based e
 diting of CSG primitives. Code and data: https://yyyyyhc.github.io/DiffCSG
 /.\n\nRegistration Category: Full Access, Full Access Supporter\n\nLanguag
 e Format: English Language\n\nSession Chair: Yonghao Yue (Aoyama Gakuin Un
 iversity)\n\n
URL:https://asia.siggraph.org/2024/program/?id=papers_466&sess=sess106
END:VEVENT
END:VCALENDAR