About

👋 Nice to meet you here!

I am currently a postdoctoral research fellow in the College of Aerospace Science and Engineering, National University of Defense Technology (NUDT), working with Prof. Qifeng Yu. Prior to this, I completed my PhD at Wuhan University (WHU), where I was fortunate to be advised by Prof. Jiayi Ma in the lab of Multi-spectral Vision Processing (MVP) here.

My research pursues modeling multi-modal and multi-dimensional data and developing cutting-edge solutions to advance next-generation intelligent perception. It primarily unfolds on two complementary levels: (a) computational imaging fundamentals; (b) application-oriented projects. I have conducted research in the intersection of machine learning and remote sensing science, and have developed a series of methods for image synthesis, image restoration, and pixel-level classification. These methods have been preliminarily applied in aerospace, earth observations, smart cities, etc. I am devoted to developing innovative AI solutions for real-world challenges, and I hope to contribute to a better understanding of the world around us.

🍵 I'm consistently open to forging new collaborations —— If you are interested in my research or would like to have a casual chat, please do not hesitate to reach me.

📢 BTW, I am delighted to assist undergraduate students in their academic and personal growth. Whether you need guidance on course selection, advice on research, or support navigating the challenges of university life, I am available to help. Please feel free to ask any questions you may have, and I will offer my support to the best of my ability.

Research

Multi-modal Fusion-based Hyperspectral Image SynthesisOngoing

Acquiring and processing hyperspectral data can be time-consuming and expensive, which causes a series of issues involving data scarcity, class skew, and target background restriction. It hinders the progress of large-scale AI-based studies and a range of applications.
To this end, we tend to raise a new paradigm of HSI synthesis, generating a vast amount of HSI with a rich diversity in various categories and scenes, closely resembling realistic data.

Robust Hyperspectral Image Restoration for Complex DegradationOngoing

Hyperspectral images are often affected by noise due to unstable factors in the complex imaging chain. This noise can significantly degrade the content and visual quality of the images, hindering their analysis and interpretation.
To alleviate these issues, we take advantage of the inherent characteristics of hyperspectral images and unique properties of hyperspectral degradation and incorporate both data-driven and model-driven strategies to propose a series of denoising and destriping methods.

Model-based Deep Learning Hyperspectral Image ClassificationProspective

A hyperspectral image is a three-dimensional data cube with high dimensionality, a strong correlation between adjacent bands, a highly nonlinear data structure, and few training samples. These characteristics make the hyperspectral image classification task challenging.
To tackle these challenges, we have endeavored to develop methods that improve classification accuracy, efficiency, and generalization ability.

Publications

This includes some of my publications. Check Google Scholar for a more complete list.

  • Unmixing before Fusion: A Generalized Paradigm for Multi-Source-based Hyperspectral Image Synthesis
    URL | PDF | Poster | Slides
    Yang Yu*, Erting Pan*, Xinya Wang, Yuheng Wu, Xiaoguang Mei, and Jiayi Ma
    IEEE/CVF International Conference on Computer Vison and Pattern Recognition (CVPR), Seattle, WA, USA, Jun.2024.
  • From the Abundance Perspective: Multi-modal Fusion-based Hyperspectral Image Synthesis URL | PDF | CODE
    Erting Pan*, Yang Yu*, Xiaoguang Mei, Jun Huang, and Jiayi Ma
    Information Fusion
  • Hyperspectral Image Destriping and Denoising from a Task Decomposition View URL | PDF | CODE
    Erting Pan, Ma Yong, Xiaoguang Mei, Fan Fan, and Jiayi Ma
    Pattern Recognition(PR), 2023.
  • Hyperspectral Image Denoising via Spectral Noise Distribution Bootstrap URL | PDF | CODE
    Erting Pan, Ma Yong, Xiaoguang Mei, Fan Fan, and Jiayi Ma
    Pattern Recognition(PR), 2023.
  • Progressive Hyperspectral Image Destriping with An Adaptive Frequencial Focus URL | PDF | CODE
    Erting Pan, Ma Yong, Xiaoguang Mei, Fan Fan, Jun Huang, and Jiayi Ma
    IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2023.
  • D2Net: Deep Denoising Network in Frequency Domain for Hyperspectral Image URL | PDF | CODE
    Erting Pan, Ma Yong, Xiaoguang Mei, Jun Huang, Fan Fan, and Jiayi Ma
    IEEE/CAA Journal of Automatica Sinica, 2023.
  • SQAD: Spatial-Spectral Quasi-Attention Recurrent Network for Hyperspectral Image Denoising URL | PDF | CODE
    Erting Pan, Ma Yong, Xiaoguang Mei, Fan Fan, Jun Huang, and Jiayi Ma
    IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2022.

Editorial Activities

Guest Editor

    Geo-Spatial Information Science

Reviewer

    IEEE/CVF Conference on Computer Vision and Pattern Recognition
    Annual AAAI Conference on Artificial Intelligence
    IEEE Transactions on Pattern Analysis and Machine Intelligence
    IEEE Transactions on Image Processing
    IEEE Geoscience and Remote Sensing Magazine
    IEEE/CAA Journal of Automatica Sinica
    IEEE Transactions on Neural Networks and Learning Systems
    IEEE Transactions on Cybernetics,
    IEEE Transactions on Circuits and Systems for Video Technology
    Information Fusion
    IEEE Transactions on Geoscience and Remote Sensing
    IEEE Transactions on Multimedia
    Neural Networks