Cheng Sun

About me

I am a Ph.D. candidate at the Information Systems and Applications at National Tsing Hua University, working with Prof. Hwann-Tzong Chen (at VLLab) and Prof. Min Sun (at VSLab). My research interests encompass wide application of computer vision, such as 360 panorama, scene modeling and understanding, object detection. Currently, I'm working on multiviews 3D reconstruction.

News

Publications

Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstruction

Cheng Sun, Min Sun, Hwann-Tzong Chen

Technical report, Nov. 2021

[Paper] [Codes]


Specialize and Fuse: Pyramid Output Representation for Semantic Segmentation

Chi-Wei Hsiao*, Cheng Sun*, Hwann-Tzong Chen, Min Sun

International Conference on Computer Vision (ICCV), 2021

[Paper]


HoHoNet: 360 Indoor Holistic Understanding with Latent Horizontal Features

Cheng Sun, Min Sun, Hwann-Tzong Chen

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021

[Paper] [Codes] [Project Page] [Video]


Indoor Panorama Planar 3D Reconstruction via Divide and Conquer

Cheng Sun, Chi-Wei Hsiao, Ning-Hsu Wang , Min Sun, Hwann-Tzong Chen

IEEE Conference on Computer Vision and Pattern Recognition (CVPR Oral), 2021

[Paper] [Codes] [Video]


Leveraging Auxiliary Information from EMR for Weakly Supervised Pulmonary Nodule Detection

Hao-Hsiang Yang, Fu-En Wang, Cheng Sun, Kuan-Chih Huang, Hung-Wei Chen, Yi Chen, Hung-Chih Chen, Chun-Yu Liao, Shih-Hsuan Kao, Yu-Chiang Frank Wang, Chou-Jin Lan

International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2021

[Paper]


360-Indoor: Towards Learning Real-World Objects in 360° Indoor Equirectangular Images

Shih-Han Chou, Cheng Sun, Wen-Yen Chang, Wan-Ting Hsu, Min Sun, Jianlong Fu

IEEE Winter Conference on Applications of Computer Vision (WACV), 2020

[Paper] [Dataset] [Project Page]


Flat2Layout: Flat Representation for Estimating Layout of General Room Types

Chi-Wei Hsiao, Cheng Sun, Min Sun, Hwann-Tzong Chen

[Paper]


HorizonNet: Learning Room Layout with 1D Representation and Pano Stretch Data Augmentation

Cheng Sun, Chi-Wei Hsiao, Min Sun, Hwann-Tzong Chen

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019

[Paper] [Codes] [Project Page]

Invited spotlight presentation at 3D Scene Generation Workshop.
Invited student presentation at Augmented Intelligence and Interaction (AII) Workshop.


DLWV2: a Deep Learning-based Wearable Vision-system with Vibrotactile-feedback for Visually Impaired People to Reach Objects

Meng-Li Shih, Yi-Chun Chen, Chia-Yu Tung, Cheng Sun, Ching-Ju Cheng, Liwei Chan, Srenivas Varadarajan, Min Sun

IEEE/RSJ Conference on Intelligent Robots and Systems (IROS), 2018

[Paper]

Side Projects

py360convert
  • Pure python w/ numpy/scipy implmentation
  • Convertion between equirectangular and cubemap
  • Convertion from equirectangular to perspective view
Yet another SAT solver
  • 2000+ lines C/C++ codes
  • Features - Iterative DPLL, 2 Literal Watching, Nonchronological Backtracking, Learnt Clause Minimization...

Experience

PhD candidate at Information Systems and Applications at National Tsing Hua University

2020 Sep - Now

Working with Prof. Hwann-Tzong Chen (at VLLab) and Prof. Min Sun (at VSLab).

M.S Degree in Information Systems and Applications at National Tsing Hua University

2018 Sep - 2020 Aug

GPA – 4.28 (0~4.3 scale)

B.S Degree in Computer Science at National Chiao Tung University

2014 Sep - 2018 Feb

GPA – 87.66 (0~100 scale, ranked 3rd in class)

SINICA Summer Internship

2017 July - 2017 August

Taiwan Open Source Conference Frontend Engineer Volunteer

2015 July - 2017 March

Implement website for SITCON CAMP 2015, SITCON 2016, SITCON 2017 and COSCUP 2016.

Miscellaneous

I were enthusiastic about GO game and played until senior high school, achieving amateur 2 dan. Currently I only play with leela-zero sometimes when deubgging model.


I played ACM International Collegiate Programming Contest (ACM-ICPC) and competitive programming when I was undergrade. The competitors are asked to solve 5 to 13 problems in 3 to 5 hours using algorithm, data structure and their problem solving skill. Our team use C/C++ and achieved 8th place in 2016 Taiwan regional and 9th Place in 2016 Korean Regional. Add me as codeforces friend.