My name is Zixiang Zhao (赵子祥, pronounced ‘Tzu-Hsiang Chao’). I am currently a Postdoctoral Researcher at Photogrammetry and Remote Sensing Group, ETH Zürich, Switzerland, supervised by Prof. Konrad Schindler.

I received my Ph.D. degree in Statistics from School of Mathematics and Statistics, Xi’an Jiaotong University, supervised by Prof. Jiangshe Zhang and worked closely with Prof. Deyu Meng and Prof. Shuang Xu.

Previously, I was a visiting Ph.D. student at Computer Vision Lab, ETH Zürich, Switzerland, supervised by Prof. Luc Van Gool and worked closely with Prof. Radu Timofte, Dr. Yulun Zhang and Dr. Kai Zhang.

Additionally, I worked as a research assistant at Visual Computing Group, Harvard University, USA, supervised by Prof. Hanspeter Pfister and collaborated closely with Dr. Zudi Lin. I also contributed to Multimedia Analytics Laboratory, City University of Hong Kong, Hong Kong, supervised by Prof. Kede Ma.

My research lies at low-level computer vision and model-based image processing, and my current research is multi-modal image fusion and restoration.

I’m open to any kinds of collaboration. Please feel free to contact me directly through email.

Email:
zixiang.zhao[AT]hotmail[DOT]com
zixiang.zhao[AT]ethz[DOT]ch
zixiangzhao[AT]stu[DOT]xjtu[DOT]edu[DOT]cn (Soon to be deprecated)

CURRENT RESEARCH

  1. Low-Level Vision
    • Image & Video Restoration and Enhancement
    • Image Fusion (infrared-visible, remote sensing, multi-focus, multi-exposure images)
    • Multi-modal Image Restoration
  2. Computer Vision
    • Generative Models (e.g., denoising diffusion models, generative adversarial networks)
    • Domain Adaption
  3. Machine Learning
    • Self-supervised learning
    • Clustering

NEWS

  • [2025-03] One paper on Unpaired Point Cloud Completion is accepted by IEEE TMM.
  • [2025-02] One paper on multi-modality image fusion is accepted by CVPR 2025.
  • [2025-01] One paper on weight binarization for diffusion model is accepted by ICLR 2025.
  • [2024-11] One paper on multi-modality image fusion is accepted by IEEE TCSVT.
  • [2024-09] One paper on image fusion is accepted by IJCV; One paper on continual learning is accepted by NeurIPS 2024; One paper on hyper-spectral image denoising is accepted by IEEE TGRS.
  • [2024-06] One paper on seismic data processing is accepted by IEEE TGRS.
  • [2024-05] Two papers are accepted by ICML 2024 (One first-authored paper on vision-language-based image fusion and the other on residual binarization for image super-resolution).
  • [2024-04] Gave a presentation @ Peking University.
  • [2024-02] One first-authored paper on multi-modality image fusion is accepted by CVPR 2024.
  • [2023-10] One paper on 3D Object Recognition is accepted by IEEE TNNLS.
  • [2023-09] Gave a presentation @ Cambridge University.
  • [2023-07] Two first-authored papers on image fusion and depth map super-resolution are accepted by ICCV 2023 (1 Oral & 1 Poster).
  • [2023-05] Named one of the CVPR 2023 Outstanding Reviewers.
  • [2023-03] One first-authored paper on multi-modality image fusion is accepted by CVPR 2023.
  • [2022-06] Two papers on seismic data processing are accepted by IEEE TGRS.
  • [2022-03] One first-authored paper on guided depth map super-resolution is accepted by CVPR 2022 (Oral) .
  • [2021-04] One first-authored paper on model-driven image fusion is accepted by IEEE TCSVT.
  • [2021-03] Two papers (one first-authored) on remote sensing image processing are accepted by ICME 2021 (1 Oral & 1 Poster).
  • [2021-03] One paper on remote sensing image processing is accepted by CVPR 2021.
  • [2020-07] One first-authored paper on Bayesian image fusion is accepted by Signal Processing.
  • [2020-04] One first-authored paper on data-driven image fusion is accepted by IJCAI 2020.

PUBLICATIONS

You can find the full list on my Google Scholar.

Selected Publications

  1. Image Fusion via Vision-Language Model.
    • Zixiang Zhao, Lilun Deng, Haowen Bai, Yukun Cui, Zhipeng Zhang, Yulun Zhang, Haotong Qin, Dongdong Chen, Jiangshe Zhang, Peng Wang, Luc Van Gool
    • Accepted by ICML 2024
    • Introduce a novel fusion paradigm, for the first time, utilizing explicit textual information to guide image fusion
      [Project Page], [Paper], [ArXiv], [Code]
  2. Equivariant Multi-Modality Image Fusion.
    • Zixiang Zhao, Haowen Bai, Jiangshe Zhang, Yulun Zhang, Kai Zhang, Shuang Xu, Dongdong Chen, Radu Timofte, Luc Van Gool
    • Accepted by CVPR 2024
    • Propose a novel end-to-end self-supervised fusion algorithm based on the equivariant sensing and imaging prior
      [Paper], [ArXiv], [Code]
  3. DDFM: Denoising Diffusion Model for Multi-Modality Image Fusion.
    • Zixiang Zhao, Haowen Bai, Yuanzhi Zhu, Jiangshe Zhang, Shuang Xu, Yulun Zhang, Kai Zhang, Deyu Meng, Radu Timofte, Luc Van Gool
    • Accepted by ICCV 2023 (ORAL)
    • Propose a novel fusion algorithm based on the denoising diffusion sampling model
      [Paper], [ArXiv], [Code]
  4. Spherical Space Feature Decomposition for Guided Depth Map Super-Resolution.
    • Zixiang Zhao, Jiangshe Zhang, Xiang Gu, Chengli Tan, Shuang Xu, Yulun Zhang, Radu Timofte, Luc Van Gool
    • Accepted by ICCV 2023
    • Propose a Spherical Space feature Decomposition network (SSDNet) and spherical contrast refinement for guided depth super-resolution
      [Paper], [ArXiv], [Code]
  5. CDDFuse: Correlation-Driven Dual-Branch Feature Decomposition for Multi-Modality Image Fusion.
    • Zixiang Zhao, Haowen Bai, Jiangshe Zhang, Yulun Zhang, Shuang Xu, Zudi Lin, Radu Timofte, Luc Van Gool
    • Accepted by CVPR 2023
    • Propose a Correlation-Driven feature Decomposition Fusion (CDDFuse) network for multi-modality image fusion
      [Paper], [ArXiv], [Code]
  6. Discrete Cosine Transform Network for Guided Depth Map Super-Resolution.
    • Zixiang Zhao, Jiangshe Zhang, Shuang Xu, Zudi Lin, Hanspeter Pfister
    • Accepted by CVPR 2022 (ORAL)
    • Propose an advanced Discrete Cosine Transform Network (DCTNet) for guided depth super-resolution
      [Paper], [ArXiv], [Code]
  7. Deep Convolutional Sparse Coding Networks for Interpretable Image Fusion.
    • Zixiang Zhao, Jiangshe Zhang, Haowen Bai, Yicheng Wang, Yukun Cui, Lilun Deng, Kai Sun, Chunxia Zhang, Junmin Liu, Shuang Xu
    • Accepted by CVPR Workshop 2023
    • Gave three deep convolutional sparse coding networks for interpretable image fusion via unfolding the iterative shrinkage and thresholding algorithm
      [Paper], [Code]
  8. Efficient and Model-Based Infrared and Visible Image Fusion via Algorithm Unrolling.
    • Zixiang Zhao, Shuang Xu, Jiangshe Zhang, Junmin Liu, Chunxia Zhang, Junmin Liu
    • Accepted by IEEE Transactions on Circuits and Systems for Video Technology (IEEE TCSVT) 2021 (Top 1% Highly Cited Paper)
    • Presented an algorithm unrolling based interpretable deep image decomposition network for infrared and visible image fusion
      [Paper], [Arxiv], [Code]
  9. Task-driven Image Fusion with Learnable Fusion Loss.
    • Haowen Bai, Jiangshe Zhang✉, Zixiang Zhao✉, Yichen Wu, Lilun Deng, Yukun Cui, Shuang Xu
    • Accepted by CVPR 2025
    • Propose a fusion framework incorporating a learnable fusion loss guided by downstream task loss
      [Paper], [Arxiv], [Code]
  10. BinaryDM: Accurate Weight Binarization for Efficient Diffusion Models.
    • Xingyu Zheng, Xianglong Liu, Haotong Qin, Xudong Ma, Mingyuan Zhang, Haojie Hao, Jiakai Wang, Zixiang Zhao, Jinyang Guo, Michele Magno
    • Accepted by International Conference on Learning Representations (ICLR) 2025
    • Propose a weight binarization approach for diffusion models.
      [Paper], [Arxiv], [Code]
  11. RefComp: A Reference-guided Unified Framework for Unpaired Point Cloud Completion.
    • Yixuan Yang, Jinyu Yang, Zixiang Zhao, Victor Sanchez, Feng Zheng
    • Accepted by IEEE Transactions on Multimedia (IEEE TMM) 2025
    • Propose a unified framework for unpaired point cloud completion
      [Paper], [Arxiv], [Code]
  12. Make Continual Learning Stronger via C-Flat.
    • Ang Bian, Wei Li, Hangjie Yuan, Yu Chengrong, Mang Wang, Zixiang Zhao, Aojun Lu, Pengliang Ji, Tao Feng
    • Accepted by Neural Information Processing Systems (NeurIPS) 2024
    • Propose a Continual Flatness method featuring a flatter loss landscape tailored for continual learning.
      [Paper], [Arxiv], [Code]
  13. Flexible Residual Binarization for Image Super-Resolution.
    • Yulun Zhang, Haotong Qin, Zixiang Zhao, Xianglong Liu, Martin Danelljan, Fisher Yu
    • Accepted by International Conference on Machine Learning (ICML) 2024
    • Proposed a flexible residual binarization image super-resolution method with second-order residual binarization and distillation-guided training.
      [Paper], [Arxiv], [Code]
  14. ReFusion: Learning Image Fusion from Reconstruction with Learnable Loss via Meta-Learning.
    • Haowen Bai, Zixiang Zhao✉, Jiangshe Zhang✉, Yichen Wu, Lilun Deng, Yukun Cui, Shuang Xu, Baisong Jiang
    • Accepted by International Journal of Computer Vision (IJCV) 2024
    • Propose a unified meta-learning based image fusion framework that dynamically optimizes the fusion loss through source image reconstruction
      [Paper], [Arxiv], [Code]
  15. Deep Unfolding Multi-modal Image Fusion Network via Attribution Analysis.
    • Haowen Bai, Zixiang Zhao, Jiangshe Zhang, Baisong Jiang, Lilun Deng, Yukun Cui, Shuang Xu, Chunxia Zhang
    • Accepted by IEEE Transactions on Circuits and Systems for Video Technology (IEEE TCSVT) 2024
    • Employed attribution analysis to tailor fused images more effectively for semantic segmentation and enhance the fusion-segmentation interaction
      [Paper], [Arxiv], [Code]
  16. Pan-Denoising: Guided Hyperspectral Image Denoising via Weighted Represent Coefficient Total Variation.
    • Shuang Xu, Qiao Ke, Jiangjun Peng, Xiangyong Cao, Zixiang Zhao
    • Accepted by IEEE Transactions on Geoscience and Remote Sensing (IEEE TGRS) 2024
    • hyperspectral image denoising using panchromatic image guidance
      [Paper], [Arxiv], [Code]
  17. Simultaneous Automatic Picking and Manual Picking Refinement for First-Break.
    • Haowen Bai, Zixiang Zhao, Jiangshe Zhang, Yukun Cui, Chunxia Zhang, Zhenbo Guo, Yongjun Wang
    • Accepted by IEEE Transactions on Geoscience and Remote Sensing (IEEE TGRS) 2024
    • Presented an algorithm unrolling based interpretable deep image decomposition network for infrared and visible image fusion
      [Paper], [Arxiv], [Code]
  18. CACNN: Capsule Attention Convolutional Neural Networks for 3D Object Recognition.
    • Kai Sun, Jiangshe Zhang, Shuang Xu, Zixiang Zhao, Chunxia Zhang, Junmin Liu, Junying Hu
    • Accepted by IEEE Transactions on Neural Networks and Learning Systems (IEEE TNNLS) 2023
    • Presented an algorithm unrolling based interpretable deep image decomposition network for infrared and visible image fusion
      [Paper], [Arxiv], [Code]
  19. Hybrid Loss Guided Coarse-to-fine Model for Seismic Data Consecutively Missing Trace Reconstruction.
    • Xiaoli Wei, Chunxia Zhang, Hongtao Wang, Zixiang Zhao, Deng Xiong, Shuang Xu, Jiangshe Zhang, Sang-Woon Kim
    • Accepted by IEEE Transactions on Geoscience and Remote Sensing (IEEE TGRS) 2022
    • Propose a coarse-to-fine seismic data consecutively missing traces interpolation
      [Paper], [Arxiv], [Code]
  20. Automatic Velocity Picking Using a Multi-Information Fusion Deep Semantic Segmentation Network.
    • Hongtao Wang, Jiangshe Zhang, Zixiang Zhao, Chunxia Zhang, Long li, Zhiyu Yang, Weifeng Geng
    • Accepted by IEEE Transactions on Geoscience and Remote Sensing (IEEE TGRS) 2022
    • Propose a multi-information fusion network to estimate stacking velocity from the fusion information of velocity spectra
      [Paper], [Arxiv], [Code]
  21. Deep Gradient Projection Networks for Pan-sharpening.
    • Shuang Xu, Jiangshe Zhang, Zixiang Zhao, Kai Sun, Junmin Liu, Chunxia Zhang
    • Accepted by CVPR 2021
    • Develop a model-based deep pan-sharpening approach via Deep Gradient Projection Network
      [Paper], [ArXiv], [Code]
  22. FGF-GAN: A Lightweight Generative Adversarial Network for Pansharpening via Fast Guided Filter.
    • Zixiang Zhao, Jiangshe Zhang, Shuang Xu, Kai Sun, Lu Huang, Junmin Liu, Chunxia Zhang
    • Accepted by ICME 2021 (ORAL)
    • Propose a generative adversarial network for pansharpening via the fast guided filter and the spatial attention module
      [Paper], [ArXiv], [Code]
  23. DIDFuse: Deep Image Decomposition for Infrared and Visible Image Fusion.
    • Zixiang Zhao*, Shuang Xu*, Chunxia Zhang, Junmin Liu, Jiangshe Zhang
    • Accepted by IJCAI 2020
    • Proposed a data-driven auto-encoder based network to accomplish the two-scale decomposition for image fusion
      [Paper], [ArXiv], [Code], [Accepted list]
  24. Bayesian Fusion for Infrared and Visible Images.
    • Zixiang Zhao, Shuang Xu, Chunxia Zhang, Junmin Liu, Jiangshe Zhang
    • Accepted by Signal Processing 2020
    • Established a Bayesian fusion model with a hierarchical Bayesian manner and the total-variation penalty, which can be inferred by the EM algorithm
      [Paper], [ArXiv], [Code]

Professional Services

PC Member | Reviewer

Conferences

  • IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • IEEE International Conference on Computer Vision (ICCV)
  • European Conference on Computer Vision (ECCV)
  • Conference on Neural Information Processing Systems (NeurIPS)
  • International Conference on Learning Representations (ICLR)
  • ACM SIGGRAPH
  • International Joint Conference on Artificial Intelligence (IJCAI)
  • AAAI Conference on Artificial Intelligence (AAAI)
  • IEEE International Conference on Robotics and Automation (ICRA)
  • Asian Conference on Computer Vision (ACCV)

Journals

  • IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
  • International Journal of Computer Vision (IJCV)
  • IEEE Transactions on Image Processing (TIP)
  • IEEE Transactions on Neural Networks and Learning Systems (TNNLS)
  • IEEE Transactions on Multimedia (TMM)
  • IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)
  • IEEE Transactions on Computational Imaging (TCI)
  • IEEE Transactions on Knowledge and Data Engineering (TKDE)
  • Information Fusion
  • Pattern Recognition
  • Signal Processing
  • IEEE Signal Processing Letters (SPL)

Membership

  • IEEE Graduate Student Member
  • VALSE (Vision and Learning SEminar) Student Club Member
  • AI TIME PhD Branch Leader

Selected Talks

  • CVPR Shenzhen Pre-Conference Meeting.
    at: [04/2024] Peking University, Shenzhen, China

  • Prior Knowledge-Guided Multi-Modal Image Fusion.
    at: [02/2024] University of Electronic Science and Technology of China (UESTC), Online

  • Conference on Frontiers in Mathematics Doctoral Student: Prior Knowledge-Guided Multi-Modal Image Fusion.
    at: [10/2023] Peking University, Beijing, China

  • Multi-modal Image Fusion and Image Reconstruction [Website].
    at: [09/2023] Cambridge University, Online

  • VALSE (Vision and Learning SEminar) 2022: Low-level Vision Session Student Spotlight Talk [Website].
    at: [08/2022] Tianjin, China

  • AI TIME PhD Presentation (CVPR 2022): Discrete Cosine Transform Network for Guided Depth Map Super-Resolution [video (Chinese)].
    at: [08/2022] AI TIME & AMiner, Online

  • 2021 IJCAI-SAIA Young Elite Symposium: Deep Image Decomposition for Infrared and Visible Image Fusion [Website].
    at: [07/2021] World Artificial Intelligence Conference & Shanghai Jiao Tong University, Shanghai, China

  • AI TIME PhD Debate: Opportunities and Challenges in multimodal learning [video (Chinese)].
    at: [07/2021] AI TIME & AMiner, Beijing, China

  • AI TIME PhD Presentation (IJCAI 2020): Deep Image Decomposition for Infrared and Visible Image Fusion [video (Chinese)].
    at: [01/2021] AI TIME & AMiner, Beijing, China

More about me!

  • a die-hard fan of Kobe Bryant and Los Angeles Lakers
  • have a passionate love for basketball
  • love swimming, skiing, singing…
  • a little guitar and violin