Presentation

<
TransparentGS: Fast Inverse Rendering of Transparent Objects with Gaussians
SessionGaussian Reconstruction
DescriptionWe propose TransparentGS, a fast inverse rendering pipeline for transparent objects based on 3D-GS. The main contributions are three-fold: efficient transparent Gaussian primitives for specular refraction, GaussProbe to encode ambient light and nearby contents, and the IterQuery algorithm to reduce parallax errors in our probe-based framework.
Authors
Letian Huang
State Key Lab for Novel Software Technology, Nanjing University
Dongwei Ye
State Key Lab for Novel Software Technology, Nanjing University
Jialin Dan
State Key Lab for Novel Software Technology, Nanjing University
Chengzhi Tao
State Key Lab for Novel Software Technology, Nanjing University
Huiwen Liu
TMCC, College of Computer Science, Nankai University
Kun Zhou
State Key Lab of CAD&CG, Zhejiang University
Institute of Hangzhou Holographic Intelligent Technology
Bo Ren
TMCC, College of Computer Science, Nankai University
Yuanqi Li
State Key Lab for Novel Software Technology, Nanjing University
Yanwen Guo
State Key Lab for Novel Software Technology, Nanjing University
Jie Guo
State Key Lab for Novel Software Technology, Nanjing University
Event Type
Technical Paper
TimeWednesday, 13 August 20254:35pm - 4:45pm PDT
LocationWest Building, Rooms 301-305
ACM Digital Library Journal Papers' PDFs
Conference Papers' PDFs
Session Time & Location
Sunday, 10 August 20256:00pm - 8:45pm PDTWest Building, Ballroom AB
Wednesday, 13 August 20253:45pm - 5:15pm PDTWest Building, Rooms 301-305
Recordings
Livestreamed
Recorded
Next PresentationNext PresentationGaussian Reconstruction - Interactive Discussion
Similar Presentations
When Gaussian Meets Surfel: Ultra-fast High-fidelity Radiance Field Rendering
On-the-fly Reconstruction for Large-Scale Novel View Synthesis from Unposed Images
Image-GS: Content-Adaptive Image Representation via 2D Gaussians