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:20250110T023309Z
LOCATION:Hall B5 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241203T150800
DTEND;TZID=Asia/Tokyo:20241203T151900
UID:siggraphasia_SIGGRAPH Asia 2024_sess107_papers_922@linklings.com
SUMMARY:OLAT Gaussians for Generic Relightable Appearance Acquisition
DESCRIPTION:Technical Papers\n\nZhiyi Kuang (State Key Laboratory of CAD&C
 G, Zhejiang University); Yanchao Yang and Siyan Dong (University of Hong K
 ong); Jiayue Ma (State Key Laboratory of CAD&CG, Zhejiang University); Hon
 gbo Fu (Hong Kong University of Science and Technology); and Youyi Zheng (
 State Key Laboratory of CAD&CG, Zhejiang University)\n\nOne-light-at-a-tim
 e (OLAT) images sample a broader range of object appearance changes than i
 mages captured under constant lighting and are superior as input to object
  relighting. Although existing methods have produced reasonable relighting
  quality using OLAT images, they utilize surface-like representations, lim
 iting their capacity to model volumetric objects, such as furs. Besides, t
 heir rendering process is time-consuming and still far from being used in 
 real-time. To address these issues, we propose OLAT Gaussians to build rel
 ightable representations of objects from multi-view OLAT images. We build 
 our pipeline on 3D Gaussian Splatting (3DGS), which achieves real-time hig
 h-quality rendering. To augment 3DGS with relighting capability, we assign
  each Gaussian a learnable feature vector, serving as an index to query th
 e objects’ appearance field. Specifically, we decompose the appearance fie
 ld into light transport and scattering functions. The former accounts for 
 light transmittance and foreshortening effects, while the latter represent
 s the object’s material properties to scatter light. Rather than using an 
 off-the-shelf physically-based parametric rendering formulation, we model 
 both functions using multi-layer perceptrons (MLPs). This makes our method
  suitable for various objects, e.g., opaque surfaces, semi-transparent vol
 umes, furs, fabrics, etc. Given a camera view and a point light position, 
 we compute each Gaussian’s color as the product of the light transport val
 ue, the scattering value, and the light intensity, and then render the tar
 get image through the 3DGS rasterizer. To enhance rendering quality, we fu
 rther utilize a proxy mesh to provide OLAT Gaussians with normals to impro
 ve highlights and visibility cues to improve shadows. Extensive experiment
 s demonstrate that our method produces state-of-the-art rendering quality 
 with significantly more details in texture-rich areas than previous method
 s. Our method also achieves real-time rendering, allowing users to interac
 tively modify views and lights to get immediate rendering results, which a
 re not available from the offline rendering of previous methods.\n\nRegist
 ration Category: Full Access, Full Access Supporter\n\nLanguage Format: En
 glish Language\n\nSession Chair: Hongzhi Wu (Zhejiang University; State Ke
 y Laboratory of CAD&CG, Zhejiang University)
URL:https://asia.siggraph.org/2024/program/?id=papers_922&sess=sess107
END:VEVENT
END:VCALENDAR