LUO, Shitong 罗世通
Undergraduate
Yuanpei College, 
Peking University
Shitong Luo is a third-year undergraduate student majoring in Data Science at Yuanpei College, Peking University. He is now a member of GLab at Wangxuan Institue of Computer Technology. He is interested in geometric data processing and geometric deep learning.
Email: luost AT pku.edu.cn
GitHub
Publications
Differentiable Manifold Reconstruction for Point Cloud Denoising
3D point clouds are often perturbed by noise due to the inherent limitation of acquisition equipments, which obstructs downstream tasks such as surface reconstruction, rendering and so on. Previous works mostly infer the displacement of noisy points from the underlying surface, which however are not designated to recover the surface explicitly and may lead to sub-optimal denoising results. To this end, we propose to learn the underlying manifold of a noisy point cloud from differentiably subsampled points with trivial noise perturbation and their embedded neighborhood feature, aiming to capture intrinsic structures in point clouds. Specifically, we present an autoencoder-like neural network. The encoder learns both local and non-local feature representations of each point, and then samples points with low noise via an adaptive differentiable pooling operation. Afterwards, the decoder infers the underlying manifold by transforming each sampled point along with the embedded feature of its neighborhood to a local surface centered around the point. By resampling on the reconstructed manifold, we obtain a denoised point cloud. Further, we design an unsupervised training loss, so that our network can be trained in either an unsupervised or supervised fashion. Experiments show that our method significantly outperforms state-of-the-art denoising methods under both synthetic noise and real world noise.
Projects
Data Visualizations
Several course projects for the course Introduction to Visualization and Visual Computing.
An Implementation of Moment Shadow Mapping & SSAO
I implemented two algorithms: Moment Shadow Mapping and Screen Space Ambient Occlusion. Besides, I wrapped OpenGL's state machine style API into an object oriented API and created a reusable rendering framework.
Chinese Chess (Xiangqi) Robot
This is a course project for Introduction to Artificial Intelligence. My contribution to this project includes the fully functional prototyping robot on simulation platform Webots, the chess piece recognition program based on OpenCV and the driver program for our final real robot.