About Me

Hello! I'm a third year PhD student at CSE of Washington University in St Louis. Under the guidance of Prof. Tao Ju, my research focuses in geometric modeling. Before coming to WashU, I spent 1.5 years at The University of Iowa, under Prof. Chris Wyman . My research area then was (real-time) realistic rendering.

I obtained my M.S. degree in State Key Laboratory of Virtual Reality Tech. & Sys.s of Beihang University, Beijing, and my B.S. from Sichuan University, Chengdu, China.

I spent a great summer in 2015 at Pixar Research.


Erosion Thickness on Medial Axes of 3D Shapes (SIGGRAPH 2016) [project] [paper] [sup1] [sup2] [tool] [data]
Authors: Yajie Yan1, Kyle Sykes2, Erin Chambers2, David Letscher2 and Tao Ju1
1Washington University in St. Louis, 2St. Louis University
Join our mailing list if you are interested in medial axis!
erosion thickness Abstract: While playing a fundamental role in shape understanding, the medial axis is known to be sensitive to small boundary perturbations. Methods for pruning the medial axis are usually guided by some measure of significance. The majority of significance measures over the medial axes of 3D shapes are locally defined and hence unable to capture the scale of features. We introduce a global significance measure that generalizes in 3D the classical Erosion Thickness (ET) measure over the medial axes of 2D shapes. We give precise definition of ET in 3D, analyze its properties, and present an efficient approximation algorithm with bounded error on a piecewise linear medial axis. Experiments showed that ET outperforms local measures in differentiating small boundary noise from prominent shape features, and it is significantly faster to compute than existing global measures. We demonstrate the utility of ET in extracting clean, shape-revealing and topology-preserving skeletons of 3D shapes.

Burning the Medial Axis (SIGGRAPH 2015 poster)

Active Research

Robust medial forms of 3D geometry. (Jan. 14 - present)
   - Explore, design, and implement a suite of algorithms for computing a new global measure over the medial axis, and deriving the medial curve structure of a 3D geometry.

Past Research

Voxelizer & Sparse Octree on GPU
   - Self project.

Volumetric Photon Mapping
   - Bachelor Thesis.
volumetric Render participating media using volumetric photon mapping method. Implemented as a PBRT plugin.

Effects including:
   Single & Multi-scattering
   Volumetric caustics

Rendering Write-ups
1. PPB and Relevant Papers
2. Volumetric Radiance Estimation Comprehensive Theory
I summarized my understanding and some hard-to-follow technical details of a family of Photon Mapping methods into these write-ups.

Feature Lines Extraction on Point Models
   - Master Thesis.
feature lines Ridge(red) and Valley(blue) lines are extracted for point models under various scales. To do this, a set of quadratics surfaces are fitted, and differential geometry calculations are then carried out analytically at each point.

Applications experimented:
   compact geometry representation
   remote rendering

For details, see project page, or
Master thesis in Chinese.

Point Model Segmentation
segmentation Implemented Cohen et al.'s Variational Shape Approximation method to partition a point model into a set of regions where each region can be approximated by a plane. The partition is said to be optimal iff the sum of the error (induced by approximating a region by a plane) is the minimal among all possible partitions.


Compression of Attributes in Sparse Voxel Octree
   - Summer 2015 internship at Pixar Research

Social Media Visualization using Sankey Diagram
   - Summer 2013 project in Avaya Labs
coming soon...


22c:005 Intro to Computer Science. Fall, 2013.

22c:108 Programming for Informatics. Spring, 2013.

22c:005 Intro to Computer Science. Fall, 2012.

Best viewed in Chrome, IE9+, Safari Designed by Yajie Yan.