photo

Kang Liao 廖康

I am a fourth-year Ph.D. student at Beijing Jiaotong University (BJTU) in MePro, advised by Professor Chunyu Lin. I have been fortunate to work with Yunchao Wei at BJTU, Xiangyu Xu at Nanyang Technological University (NTU), Yulan Guo at National University of Defense Technology (NUDT), and Moncef Gabbouj at Tampere University. Recently I am conducting my research at Max Planck Institute for Informatics in Computer Graphics group advised by Vahid Babaei (Oct. 2021 - Oct. 2022).

I was borned and raised in Anqing, a picturesque city with colorful culture in China.

Research Interests

My research interest lies at the intersection of Computer Vision and Deep Learning. In particular, my work focuses on image generation/restoration, multiple view geometry, 3D vision and point cloud, and 360° panoramic scene understanding, etc.

  kang_liao AT bjtu DOT edu DOT cn             ZHIHU             Google Scholar

Publications

Preprint

Cylin-Painting: Seamless 360° Panoramic Image Outpainting and Beyond with Cylinder-Style Convolutions
Kang Liao, Xiangyu Xu, Chunyu Lin, Wenqi Ren, Yunchao Wei, Yao Zhao
[arXiv] [Github]

FisheyeEX: Polar Outpainting for Extending the FoV of Fisheye Lens
Kang Liao, Chunyu Lin, Yunchao Wei, Yao Zhao
[arXiv]

Towards Reliable Image Outpainting: Learning Structure-Aware Multimodal Fusion with Depth Guidance
Lei Zhang, Kang Liao, Chunyu Lin, Yao Zhao
[arXiv]

Deep Rotation Correction without Angle Prior
Lang Nie, Chunyu Lin, Kang Liao, Shuaicheng Liu, Yao Zhao
[arXiv] [Github]

FishFormer: Annulus Slicing-based Transformer for Fisheye Rectification with Efficacy Domain Exploration
Shangrong Yang, Chunyu Lin, Kang Liao, Yao Zhao
[arXiv]

Complementary Bi-directional Feature Compression for Indoor 360° Semantic Segmentation with Self-distillation
Zishuo Zheng, Chunyu Lin, Lang Nie, Kang Liao, Zhijie Shen, Yao Zhao
[arXiv]

2022

PanoFormer: Panorama Transformer for Indoor 360° Depth Estimation
Zhijie Shen, Chunyu Lin, Kang Liao, Lang Nie, Zishuo Zheng, Yao Zhao
European Conference on Computer Vision (ECCV), 2022
[arXiv] [Github]

Deep Rectangling for Image Stitching: A Learning Baseline
Lang Nie, Chunyu Lin, Kang Liao, Shuaicheng Liu, Yao Zhao
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR, Oral), 2022
[PDF] [Github] [Chinese Blog]

SivsFormer: Parallax-Aware Transformers for Single-image-based View Synthesis
Chunlan Zhang, Chunyu Lin, Kang Liao, Lang Nie, Yao Zhao
IEEE Conference on Virtual Reality and 3D User Interfaces (IEEE VR), 2022
[PDF]

2021

Multi-Level Curriculum for Training A Distortion-Aware Barrel Distortion Rectification Model
Kang Liao, Chunyu Lin, Lixin Liao, Yao Zhao, Weiyao Lin
International Conference on Computer Vision (ICCV), 2021
[PDF]

Towards Complete Scene and Regular Shape for Distortion Rectification by Curve-Aware Extrapolation
Kang Liao, Chunyu Lin, Yunchao Wei, Feng Li, Shangrong Yang, Yao Zhao
International Conference on Computer Vision (ICCV), 2021
[PDF]

Appearance Flow-based Progressively Complementary Network for Distortion Rectification
Shangrong Yang*, Chunyu Lin*, Kang Liao*, Chunjie Zhang, Yao Zhao
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021
(*Equal contribution. Personal efforts: idea - 50%, paper revision - 30%, rebuttal - 90%)
[PDF] [Github]

A Deep Ordinal Distortion Estimation Approach for Distortion Rectification
Kang Liao, Chunyu Lin, Yao Zhao
IEEE Transactions on Image Processing (IEEE TIP), 2021
[PDF]

Pseudo-LiDAR Point Cloud Interpolation Based on 3D Motion Representation and Spatial Supervision
Haojie Liu*, Kang Liao*, Chunyu Lin, Yao Zhao, Yulan Guo
IEEE Transactions on Intelligent Transportation Systems (IEEE TITS), 2021
(*Equal contribution. Personal efforts: idea - 90%, paper writing - 40%, rebuttal - 90%)
[PDF] [Github]

Unsupervised Deep Image Stitching: Reconstructing Stitched Features to Images
Lang Nie, Chunyu Lin, Kang Liao, Shuaicheng Liu, Yao Zhao
IEEE Transactions on Image Processing (IEEE TIP), 2021
[PDF] [Github] [Chinese Blog]

Depth-Aware Multi-Grid Deep Homography Estimation with Contextual Correlation
Lang Nie, Chunyu Lin, Kang Liao, Shuaicheng Liu, Yao Zhao
IEEE Transactions on Circuits and Systems for Video Technology (IEEE TCSVT), 2021
[PDF] [Github]

Revisiting Radial Distortion Rectification in Polar-coordinates: A New and Efficient Learning Perspective
Keyao Zhao, Chunyu Lin, Kang Liao, Shangrong Yang, Yao Zhao
IEEE Transactions on Circuits and Systems for Video Technology (IEEE TCSVT), 2021
[PDF]

Joint Distortion Rectification and Super-Resolution for Self-Driving Scene Perception
Keyao Zhao, Kang Liao, Chunyu Lin, Yao Zhao
Neurocomputing, 2021
[PDF]

Pseudo-LiDAR Point Cloud Magnification
Chunlan Zhang, Kang Liao, Chunyu Lin, Yao Zhao
Neurocomputing, 2021
[PDF]

2020

Model-Free Distortion Rectification Framework Bridged by Distortion Distribution Map
Kang Liao, Chunyu Lin, Yao Zhao, Mai Xu
IEEE Transactions on Image Processing (IEEE TIP), 2020
[PDF]

OIDC-Net: Omnidirectional Image Distortion Correction via Coarse-to-fine Region Attention
Kang Liao, Chunyu Lin, Yao Zhao, Moncef Gabbouj
IEEE Journal of Selected Topics in Signal Processing (IEEE JSTSP), 2020
[PDF]

A view-free image stitching network based on global homography
Lang Nie, Chunyu Lin, Kang Liao, Meiqin Liu, Yao Zhao
Journal of Visual Communication and Image Representation (JVCIR), 2020
[PDF] [Github]

Panoramic Image Quality-Enhancement by Fusing Neural Textures of the Adaptive Initial Viewport
Shiyuan Li, Chunyu Lin, Kang Liao, Yao Zhao, Xue Zhang
IEEE Virtual Reality Workshop (IEEE VR), 2020
[PDF]

2019

Distortion Rectification from Static to Dynamic: A Distortion Sequence Construction Perspective
Kang Liao, Chunyu Lin, Yao Zhao, Moncef Gabbouj
IEEE Transactions on Circuits and Systems for Video Technology (IEEE TCSVT), 2019
[PDF]

DR-GAN: Automatic Radial Distortion Rectification Using Conditional GAN in Real-Time
Kang Liao, Chunyu Lin, Yao Zhao, Moncef Gabbouj
IEEE Transactions on Circuits and Systems for Video Technology (IEEE TCSVT), 2019
[PDF] [Github]


Selected Award

  • Second Price in China Graduate AI Innovation Competition (Dec. 2021. Team leader, 2% awarded <30/1505>.)
  • ZHIXING Scholarship (or President Scholarship) of BJTU (Jun. 2021, the greatest honor for the graduate students at BJTU. 1‰ awarded.)
  • China National Scholarship (Oct. 2021, Oct. 2020. 2% awarded.)
  • China Scholarship Council (CSC) Scholarship (Aug. 2020)
  • PhD Innovation Fund of BJTU (May. 2020)
  • First-class PhD academic scholarship of BJTU (Oct. 2019, Oct. 2018)
  • First-class graduate students academic scholarship of BJTU (Oct. 2017)
  • First Prize in Shaanxi Province Programming Competition (Jun. 2015)

Professional Activities

Reviewer

  • International Journal of Computer Vision (IJCV)
  • IEEE Transactions on Image Processing (TIP)
  • IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2022
  • European Conference on Computer Vision (ECCV) 2022
  • International Conference on Computer Vision (ICCV) 2021
  • International Conference on Multimedia and Expo (ICME) 2021
  • IEEE Winter Conference on Applications of Computer Vision (WACV) 2022

Talk

  • "Towards Complete Scene and Regular Shape for Distortion Rectification by Curve-Aware Extrapolation", Graduate academic forum of Chinese journal of image and graphics, Sep. 2021
  • "Wide Field of View (FoV) Computer Vision", UT Austin, Apr. 2022
  • "Wide Field of View (FoV) Computer Vision and Beyond", BJTU, May. 2022

Blog