教师简介

魏朋旭，博士，副教授。博士毕业于中国科学院大学。已发表30余篇高水平学术论文，其中包括CVPR、ECCV等五篇会议口头报告Oral论文，一篇获PRCV最佳论文候选。构建了当前最大规模的真实图像超分辨率数据集。两次举办IEEE Seasonal School，并在领域顶会ECCV上举办真实图像超分辨率学术竞赛。组织搭建了一套系统化的Diffusion模型代码框架HCP-Diffusion(https://github.com/IrisRainbowNeko/HCP-Diffusion)。现在主要研究兴趣：底层视觉模型预训练，真实图像超分辨率数据基准构建及方法研究，图像和视频内容生成。个人主页：https://pengxuwei.github.io/

研究领域

• 1）底层视觉：面向真实世界的图像超分辨率，底层视觉预训练；

• 2）视觉内容生成：图像和视频内容生成；

• 3）模型鲁棒性及泛化性：对抗攻击与防御，模型可迁移性和泛化性

教育背景

• 中国科学院大学，计算机应用技术，博士，2018.1
• 中国矿业大学（北京），计算机科学与技术，学士，2011.6

工作经历

副教授, 中山大学, 2024-至今

副研究员，中山大学，2021-2023

博士后，中山大学，2018-2020

研究员，商汤科技，2018

主要学术兼职

审稿人：CVPR; ICCV; ECCV; ICLR; NeurIPS; IJCAI; AAAI; IEEE-TPAMI; IEEE-TIP; IEEE-TCSVT, CVIU

组织者：1）AIM: Advances in Image Manipulation Workshop, Challenge on Real Image Super-Resolution (in conjunction with ECCV2020)；2）IEEE CASS Seasonal School

科研项目

1）国家自然科学基金面上项目，主持

2）国家自然科学基金青年项目，主持

3）国家重点研发项目，子课题负责人

4）博士后面上项目，主持

5）高校业务费基本科研业务费项目，主持

6）中国科学院大学横向课题，主持

代表性论著 

• Xipeng Chen^#, Pengxu Wei^*, Liang Lin. Deductive Learning for Weakly-Supervised 3D Human Pose Estimation via Uncalibrated Cameras, AAAI Conference on Artificial Intelligence (AAAI), 2021.
• Pengxu Wei, Ziwei Xie, Hannan Lu, Zongyuan Zhan, Qixiang Ye, Wangmeng Zuo, Liang Lin^*. Component Divide-and-Conquer for Real-World Image Super-Resolution, European Conference on Computer Vision (ECCV), 2020.
• Pengxu Wei, Hannan Lu, Radu Timofte, Liang Lin, Wangmeng Zuo, et al. AIM 2020 Challenge on Real ImageSuper-Resolution: Methods and Results, European Conference on Computer Vision (ECCV) Workshop, 2020.
• RuimaoZhang^#, Wei Yang, Zhanglin Peng, Pengxu Wei^*, Xiaogang Wang, Liang Lin. Progressively Diffused Networks for Semantic Visual Parsing, Pattern Recognition (PR), 2019, 90(6): 78-86.
• Keze Wang^#, Liang Lin^*, Chenhan Jiang, Chen Qian, Pengxu Wei. 3D Human Pose Machines with Self-supervised Learning, IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 2019, 42(5):1069-1082.
• Fang Wan^#, Pengxu Wei, Jianbin Jiao, Zhenjun Han, Qixiang Ye^*. Min-Entropy Latent Model for Weakly Supervised Object Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 2019, 41(10):2395-2409.
• Pengxu Wei, Fei Qin, Fang Wan, Yi Zhu, Jianbin Jiao, Qixiang Ye. Correlated Topic Vector for Scene Classification, IEEE Transactions on Image Processing (TIP), 2017, 26(7):3221-3234.

• Fang Wan, Pengxu Wei, Jianbin Jiao, Zhenjun Han, Qixiang Ye. Min-Entropy Latent Model for Weakly Supervised Object Detection, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2018.
• Fang Wan, Pengxu Wei, Zhenjun Han, Kun Fu, Qixiang Ye. Weakly Supervised Object Detection with Correlation and Part Suppression. IEEE International Conference on Image Processing (ICIP) 2016:3638-3642.
• Xiaodan Zhang, Shenfeng He, Xinhang Song, Pengxu Wei, Shuqiang Jiang, Qixiang Ye, Jianbin Jiao, Rynson Lau. Keyword-driven Image Captioning via Context-dependent Bilateral LSTM. IEEE International Conference on Multimedia and Expo (ICME) 2017:781-786.
• Wei Ke, Yao Zhang, Pengxu Wei, Qixiang Ye, Jianbin Jiao. Pedestrian Detection via PCA Filters Based Convolutional Channel Features. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015:1394-1398.