Avatar

Zhelun Shen

Senior Researcher

Baidu


Google Scholar | Github


I am currently a senior researcher working for computer vision at Baidu. Before that, I received my master degree in signal and information processing from Peking University in 2022 and my B.E. degree, advised by Prof. Yuchao Dai, at the School of Electronics and Information, Northwestern Polytechnical University in 2019.

My earlier research centered on depth estimation, especially binocular depth estimation. Recently, I have been exploring generative AI, focusing on integrating generation and perception/reconstruction models for 3D applications.

Contact

  • shenzhelun [at] pku.edu.cn

  • (+86)-188-2179-0364

Education

  • M.E. in Signal and Information Processing

    2019 - 2022

    Peking University

  • B.E. in Communication Engineering

    2015 - 2019

    Northwestern Polytechnical University

Research Highlights

  • Publication:: 7 papers in top-tier conferences and journals with over 800 citations in total, including 3 first-authored papers (1 CVPR, 1 ECCV Oral, and 1 TPAMI). All my first-authored papers together have contributed over 600 citations.
  • Competition: Two championships in top-tier conferences’workshop challenges, i.e., CVPR2021 Argoverse Stereo Competition and ECCV2020 Robust Vision Challenge.
  • Open source: Owner of paddle’s official repo: Paddle Depth, which is a lightweight toolkit for depth information sensing, i.e., depth super resolution, depth completion, monocular depth estimation, and stereo depth estimation.

Selected Publications

  • Digging Into Uncertainty-Based Pseudo-Label for Robust Stereo Matching

    Zhelun Shen, Xibin Song, Yuchao Dai, Dingfu Zhou, Zhibo Rao, Liangjun Zhang

    IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2023

    [ Paper | Code ]

  • PCW-Net: Pyramid Combination and Warping Cost Volume for Stereo Matching

    Zhelun Shen, Yuchao Dai, Xibin Song, Zhibo Rao, Dingfu Zhou, Liangjun Zhang

    European Conference on Computer Vision (ECCV), 2022 (Oral with 2.7% acceptance rate)

    [ Paper | Code ]

  • Cfnet: Cascade and fused cost volume for robust stereo matching

    Zhelun Shen, Yuchao Dai, Zhibo Rao

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

    [ Paper | Code ]

  • Digging into Depth Priors for Outdoor Neural Radiance Fields

    Chen Wang, Jiadai Sun, Lina Liu, Chenming Wu*, Zhelun Shen*, Dayan Wu, Yuchao Dai, Liangjun Zhang

    * denotes corresponding author

    Proceedings of the 31th ACM International Conference on Multimedia (ACMMM), 2023 (Oral)

    [ Project Page | Paper | Code ]

  • Masked representation learning for domain generalized stereo matching

    Zhibo Rao, Bangshu Xiong, Mingyi He, Yuchao Dai, Renjie He, Zhelun Shen, Xing Li

    IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023

    [ Paper ]

  • TUSR-Net: Triple Unfolding Single Image Dehazing With Self-Regularization and Dual Feature to Pixel Attention

    Xibin Song, Dingfu Zhou, Wei Li, Yuchao Dai, Zhelun Shen, Liangjun Zhang, Hongdong Li

    IEEE Transactions on Image Processing (TIP), 2023

    [ Paper ]

  • Rethinking training strategy in stereo matching

    Zhibo Rao, Yuchao Dai, Zhelun Shen, Renjie He

    IEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2022

    [ Paper ]

Awards & Honors

  • 1st in Argoverse Stereo Competition (CVPR 2021 Workshop), 2021
  • 1st in stereo track of Robust Vision Challenge 2020 (ECCV2020 Workshop), 2020
  • Outstanding Undergraduate Student, Northwestern Polytechnical University, China, 2019
  • National Scholarship (Top 1% in Northwestern Polytechnical University), 2016-2017

Reviewer

  • IEEE Transactions on Image Processing (TIP)
  • IEEE Transactions on Medical Imaging (TMI)
  • Pattern Recognition (PR)
  • IEEE Robotics and Automation Letters (RAL)