Hello! My name is Haotian Zhang. I am currently a third year PhD student in computer science department at Stanford University, advised by Prof. Kayvon Fatahalian. My research interest lies in Computer Graphics and Computer Vision.

I got my bachelor's degree from computer science department, Tsinghua University and worked with Prof. Shi-Min Hu.

Publication

An Internal Learning Approach to Video Inpainting

Haotian Zhang, Long Mai, Ning Xu, Zhaowen Wang, John Collomosse, Hailin Jin ICCV 2019

We propose a novel video inpainting algorithm that simultaneously hallucinates missing appearance and motion (optical flow) information.

Website Paper Code

TextureNet: Consistent Local Parametrizations for Learning from High-Resolution Signals on Meshes

Jingwei Huang, Haotian Zhang, Li Yi, Thomas Funkhouser, Matthias Nießner, Leonidas Guibas CVPR 2019 (Oral presentation)

We introduce TextureNet, a neural network architecture designed to extract features from high-resolution signals associated with 3D surface meshes (e.g., color texture maps).

Website Paper Code

Robust Background Identification for Dynamic Video Editing

Fang-Lue Zhang, Xian Wu, Haotian Zhang, Jue Wang, Shi-Min Hu SIGGRAPH Asia 2016

We present a new, practical method that can reliably identify background features in complex video, leading to accurate camera path estimation and background layering.

Paper

Image-based Clothes Changing System

Zhao-Heng Zheng, Haotian Zhang, Fang-Lue Zhang, Tai-Jiang Mu Computational Visual Media 2017

We propose an image-based clothes changing system, exploiting body factor extraction and content-aware image warping.

Paper

Research Projects

Rekall: Specifying Video Events using Compositions of Spatiotemporal Labels

Dan Fu, Will Crichton, James Hong, Xinwei Yao, Haotian Zhang, Anh Truong, Avanika Narayan, Maneesh Agrawala, Christopher Ré, Kayvon Fatahalian

We introduce Rekall, a library that exposes a data model and programming model for compositional video event specification.

Arxiv

Eurypterid Swimming Simulation

with Jie Tan, Greg Turk

We are aimed to determine the swimming motion pattern of an extinct species -- Eurypterid. Our final result provide evidence for supporting one of the swimming motion hypothesis.

Website

Other Selected Projects

Gaseous Fluid Simulation

A gaseous fluid simulation demo. Implementation based on the algorithm proposed in Jos Stam's paper "Real-time fluid dynamics for games". Several methods are used for velocity field visualization.

Website Code

Fluid Simulation with Free Surface

A demo for fluid simulation with free surface. Implementation based on the Marker-And-Cell (MAC) method with free surface presented. Both 2D and 3D simulation are supported. Used Blobby and Marching Cubes for creating surface mesh.

Website Code

RayTrace

A rayTracer for 2D image rendering based on Phone's model. Implemented functions include diffusion, highlight, reflection, refraction, soft shadow, texture mapping, anti-alisa, KD-tree with triangle mesh rendering, random diffusion, depth of field etc.

Website Code

Mesh Simplify

A mesh simplification algorithm based on edge collapse.

Website Code