Tianwei Shen

News

* Feb, 2019 - Three papers accepted for CVPR 2019, code available for ContextDesc and R-MVSNet.

* Jan, 2019 - Check out the open-sourced DeepMatchVO for ICRA 2019.

* Sep, 2018 - Check out the open-sourced MIRRoR for ACCV 2018.

Intro

About: I am a computer vision researcher in Facebook Reality Labs. Previously, I completed my Ph.D. degree at HKUST, advised by Prof. Long Quan. I obtained my bachelor degree from Peking University, double major in machine intelligence (EECS) and psychology. As an undergraduate research intern, I spent time at UC Berkeley and worked on Captcha recognition, hosted by Prof. Dawn Song. I also developed algorithms for Altizure from Day 1.

Research interests: My research interests lie in large-scale Structure-from-Motion, SLAM and related learning problems. I aim to do research that accomplishes myself, enlightens peers and benefits the greater good.

Research

The following shows some of my Ph.D. work. Please see my google scholar for the updated list of publications.

2019

⚬ Local Feature Augmentation with Cross-Modality Context [arxiv] [code]
Zixin Luo, Tianwei Shen, Lei Zhou, Jiahui Zhang, Yao Yao, Shiwei Li, Tian Fang, Long Quan
Conference on Computer Vision and Pattern Recognition (CVPR) 2019 (Oral)

⚬ Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inference [arxiv] [code]
Yao Yao, Zixin Luo, Shiwei Li, Tianwei Shen, Tian Fang, Long Quan
Conference on Computer Vision and Pattern Recognition (CVPR) 2019

⚬ Cross-atlas Convolution for Parameterization Invariant Learning on Mesh Surface
Shiwei Li, Zixin Luo, Mingmin Zhen, Yao Yao, Tianwei Shen, Tian Fang, Long Quan
Conference on Computer Vision and Pattern Recognition (CVPR) 2019

⚬ Beyond Photometric Loss for Self-Supervised Ego-Motion Estimation [arxiv] [code]
Tianwei Shen, Zixin Luo, Lei Zhou, Hanyu Deng, Runze Zhang, Tian Fang, Long Quan
International Conference on Robotics and Automation (ICRA) 2019

2018

⚬ Matchable Image Retrieval by Learning from Surface Reconstruction [arxiv] [code][data]
Tianwei Shen^*, Zixin Luo^*, Lei Zhou, Runze Zhang, Tian Fang, Long Quan
Asian Conference on Computer Vision (ACCV) 2018

⚬ GeoDesc: Learning Local Descriptors by Integrating Geometry Constraints [pdf] [code]
Zixin Luo, Tianwei Shen, Lei Zhou, Siyu Zhu, Runze Zhang, Yao Yao, Tian Fang, Long Quan
European Conference on Computer Vision (ECCV) 2018

⚬ Learning & Matching Multi-View Descriptors for Registration of Point Clouds [pdf][code]
Lei Zhou, Siyu Zhu, Zixin Luo, Tianwei Shen, Runze Zhang, Tian Fang, Long Quan
European Conference on Computer Vision (ECCV) 2018

⚬ Distributed Very Large Scale Bundle Adjustment by Global Camera Consensus [pdf]
Runze Zhang, Siyu Zhu, Tianwei Shen, Lei Zhou, Zixin Luo, Tian Fang, Long Quan
IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI) 2018

⚬ Very Large-Scale Global SfM by Distributed Motion Averaging [pdf]
Siyu Zhu^*, Runze Zhang^*, Lei Zhou, Tianwei Shen, Tian Fang, Ping Tan, Long Quan
Conference on Computer Vision and Pattern Recognition (CVPR) 2018

2017

⚬ Progressive Large Scale-Invariant Image Matching In Scale Space [pdf] [data]
Lei Zhou, Siyu Zhu, Tianwei Shen, Jinglu Wang, Tian Fang, Long Quan
International Conference on Computer Vision (ICCV) 2017

⚬ Parallel Structure from Motion from Local Increment to Global Averaging [arxiv]
Siyu Zhu, Tianwei Shen, Lei Zhou, Runze Zhang, Jinglu Wang, Tian Fang, Long Quan
arXiv:1702.08601, 2017

2016

⚬ Color Correction for Image-Based Modeling in the Large [pdf] [supp]
Tianwei Shen, Jinglu Wang, Tian Fang, Siyu Zhu, Long Quan
Asian Conference on Computer Vision (ACCV) 2016

⚬ Graph-Based Consistent Matching for Structure-from-Motion [pdf] [supp] [datasets]
Tianwei Shen, Siyu Zhu, Tian Fang, Runze Zhang, Long Quan
European Conference on Computer Vision (ECCV) 2016

2015

⚬ Structure-driven Facade Parsing With Irregular Patterns [pdf]
Jinglu Wang, Chun Liu, Tianwei Shen, Long Quan
Asian Conference on Pattern Recognition (ACPR) 2015 (Oral)

Tutorial

☐ Tutorial: Large-Scale 3D Reconstruction from Images [website]
Organizers: Tianwei Shen, Jinglu Wang, Tian Fang, Long Quan
Asian Conference on Computer Vision (ACCV) 2016

Code

[1] DeepMatchVO - End-to-end pose and depth estimation[github]

[2] MIRorR - Matchable image retrieval [github]

[3] libvot - A C++11 multi-thread implementation of vocabulary tree [website] [github]

[4] Blender - Multi-view camera reconstruction [blog] [commits] [video]

Misc

* Technical blog (archived)

* Some handy scripts

EMAIL

ADDRESS