picture

Daeyun Shin


My research interests are in artificial intelligence at the intersection of computer vision and computer graphics. I am interested in 3D scene understanding and representation learning. I am advised by Charless Fowlkes.

Education

Since 2017, Ph.D. in Computer Science, University of California, Irvine (advisor: Charless Fowlkes)

2017, M.S. in Computer Science, University of Illinois at Urbana-Champaign (advisor: Derek Hoiem)

2015, B.S. in Computer Science, University of Illinois at Urbana-Champaign

Publications

Domain Decluttering: Simplifying Images to Mitigate Synthetic-Real Domain Shift and Improve Depth Estimation
Yunhan Zhao, Shu Kong, Daeyun Shin, Charless Fowlkes
3D Scene Reconstruction with Multi-layer Depth and Epipolar Transformers
Daeyun Shin, Zhile Ren, Erik Sudderth, Charless Fowlkes
Pixels, voxels, and views: A study of shape representations for single view 3D object shape prediction
Daeyun Shin, Charless Fowlkes, Derek Hoiem
Geometric Pose Affordance: 3D Human Pose with Scene Constraints
Zhe Wang, Liyan Chen, Shaurya Rathore, Daeyun Shin, Charless Fowlkes
Predicting Camera Viewpoint Improves Cross-dataset Generalization for 3D Human Pose Estimation
Zhe Wang, Daeyun Shin, Charless Fowlkes
3DFS: Deformable Dense Depth Fusion and Segmentation for Object Reconstruction from a Handheld Camera
Tanmay Gupta, Daeyun Shin, Naren Sivagnanadasan, Derek Hoiem
Completing 3D Object Shape from One Depth Image
Jason Rock, Tanmay Gupta, Justin Thorsen, Junyoung Gwak, Daeyun Shin, Derek Hoiem

Selected Projects

Internships

Google
Research Intern
Irvine, CA
June 2021 - Sept 2021
Snap Inc.
Research Intern
Santa Monica, CA
June 2019 - Sept 2019
Google
Software Engineering Intern
New York, NY
May 2016 - Aug 2016
Google
Software Engineering Intern
Pittsburgh, PA
May 2015 - Aug 2015
Amazon Web Services
Software Engineering Intern
Palo Alto, CA
May 2014 - Aug 2014
Amazon.com
Software Engineering Intern
Seattle, WA
May 2013 - Aug 2013

Talks

  • Berkeley Artificial Intelligence Research (BAIR) Lab, 3D Scene Reconstruction with Multi-layer Depth and Epipolar Transformers. CA, USA. Sept. 17, 2019. [slides]
  • Stanford Vision and Learning Lab (SVL) Group Meeting, 3D Scene Reconstruction with Multi-layer Depth and Epipolar Transformers. CA, USA. Sept. 16, 2019. [slides]
  • KAIST Scalable Graphics, Vision, & Robotics (SGVR) Lab, 3D Scene Reconstruction with Multi-layer Depth and Epipolar Transformers. Daejeon, Korea. Oct. 2019. [slides]
  • UCI AI/ML Seminar Series, Multi-layer Depth and Epipolar Feature Transformers for 3D Scene Reconstruction. CA, USA. Apr. 15, 2019. [page]

Professional Services

Conference Reviewer
ICLR 2021, ECCV 2020, NeurIPS 2020, CVPR 2020, ACCV 2020, ICLR 2020, AAAI 2020, ICCV 2019, CVPR 2019, WACV 2020, BMVC 2019, ACCV 2018, WACV 2019
Journal Reviewer
T-PAMI, IEEE Access

Research Topics

  • Viewer-centered 3D reconstruction.
  • Geometric output representations.
  • 3D scene understanding.
  • Learning novel objects.
  • Multi-task learning.
  • Learning from synthetic images.
  • Adaptive computation.