阚世超

讲师 硕士生导师

所在单位:计算机学院

学历:博士研究生毕业

办公地点:新校区信息楼429

联系方式:kanshichao@csu.edu.cn

学位:工学博士学位

在职信息:在职

毕业院校:北京交通大学

学科:计算机科学与技术

个人简介

2021年6月毕业于北京交通大学信息科学研究所,2019年9月至2020年10月在美国密苏里大学哥伦比亚分校访学, 2021年9月入职中南大学。主要研究方向为多模态大模型、计算机视觉、机器学习、深度学习和人工智能,在IEEE TPAMI、IEEE TIP、IEEE TCSVT等国际期刊和NeurIPS、CVPR、ACMMM、IJCAI、ECCV等国际会议上发表论文四十余篇。主持国家自然科学基金青年基金、湖南省自然科学基金青年基金项目2项。曾获2022年北京图象图形学学会优秀博士学位论文奖、OpenHW 2015 开源硬件与嵌入式计算大赛全国一等奖、2012 年全国大学生数学建模竞赛北京市一等奖等。担任IEEE Transactions on Multimedia (TMM), IEEE Transactions on Neural Network and Learning Systems (TNNLS), IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)、IEEE Transactions on Emerging Topics in Computational Intelligence (TETCI)、Neural Networks (NEUNET)、Signal Processing-Image Communication (SPIC)和Journal of Visual Communication and Image Representation (JVCIR)等期刊审稿人。担任NeurIPS、CVPR、ICCV、ICLR等会议审稿人,以及BIBM Session Chair、APBC PC Member。


欢迎具有一定编程基础,对多模态、大模型、以人工智能为主的多领域交叉感兴趣,勤奋刻苦,致力于发表高水平论文的本科生和研究生随时联系,优先欢迎有多模态大模型技术基础或者对多模态大模型感兴趣的学生准备进入课题组的研究生最好已熟悉多模态数据处理基础技术,如自然图像、医疗图像、视频、文本、语音、点云、遥感等,并熟练运用python和pytorch等深度学习框架编程,本科有多模态或大模型相关方面的科研经历或论文发表经历者优先大三和大四学生想提前接触科研可随时联系,对基础不作要求,欢迎优秀本科生提前开展科研工作,发表论文或申请专利。当前研究课题主要有:多模态大模型(Multimodal Large Language Models),开放词汇目标检索(Open Vocabulary Object Retrieval),跨模态目标检索(Cross-Modal Object Retrieval),视频目标和行为分析(Video Object and Behavior Analysis)场景图生成(Scene Graph Generation),跨模态在线增量学习(Cross-Modal Online Incremental Learning),多模态生物医学影像问答(Multimodal Biomedical Image Question-Answering等。

 

代表论文

[1] Yuming Wu, Lihui Cen, Shichao Kan*, Yongfang Xie, Multi-Layer Capsule Network with Joint Dynamic Routing for Fire Recognition [J], Image and Vision Computing (IMAVIS), 2023. (*corresponding author)

[2] Yue Zhang, Suchen Wang, Shichao Kan, Zhenyu Weng, Yigang Cen, Yap-peng Tan, POAR: Towards Open Vocabulary Pedestrian Attribute Recognition[C], ACM MM 2023. (CCF A类, oral)

[3] Lele Lv, Qing Liu, Shichao Kan, Yixiong Liang, Confidence-Aware Contrastive Learning for Semantic Segmentation[C], ACM MM 2023. (CCF A类)

[4] Yifan Wu, Shichao Kan, Min Zeng, Min Li, Singularformer: Learning to Decompose Self-Attention to Linearize the Complexity of Transformer[C], The 32nd International Joint Conference on Artificial Intelligence (IJCAI-23). (CCF A类)

[5] Shichao Kan, Zhiquan He, Yigang Cen, Yang Li, Vladimir Mladenovic, Zhihai He, Contrastive Bayesian Analysis for Deep Metric Learning [J], IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2023, 45(6): 7220-7238. (CCF A类)

[6] Shichao Kan, Yixiong Liang, Min Li, Yigang Cen, Jianxin Wang, Zhihai He, Coded Residual Transform for Generalizable Deep Metric Learning [C], Advances in Neural Information Processing Systems 35 (NeurIPS), 2022, 28601-28615. (CCF A类)

[7] Shichao Kan, Yigang Cen, Yang Li, Mladenovic Vladimir, Zhihai He, Local Semantic Correlation Modeling over Graph Neural Networks for Deep Feature Embedding and Image Retrieval [J], IEEE Transactions on Image Processing (TIP), 2022, 31:2988-3003.(CCF A类

[8] Shichao Kan#, Yue Zhang#, Fanghui Zhang, Yigang Cen, A GAN-based input-size flexibility model for single image dehazing [J], Signal Processing: Image Communication (SPIC), 2022. (CCF C类, #co-first authors

[9] Shichao Kan, Yi Cen, Yigang Cen, Mladenovic Vladimir, Yang Li, Zhihai He, Zero-Shot Learning to Index on Semantic Trees for Scalable Image Retrieval [J], IEEE Transactions on Image Processing (TIP), 2021, 30: 501-516. (CCF A类)

[10] Shichao Kan, Yigang Cen, Yang Li, Vladimir Mladenovic, Zhihai He, Relative Order Analysis and Optimization for Unsupervised Deep Metric Learning [C], Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, 13994-14003. (CCF A类

[11] Yang Li, Shichao Kan, Jianhe Yuan, Wenming Cao, Zhihai He, Spatial Assembly Networks for Image Representation Learning [C], Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, 13994-14003. (CCF A类

[12] Shichao Kan, Linna Zhang, Zhihai He, Yigang Cen, Shiming Chen, Jikun Zhou, Metric learning-based kernel transformer with triplets and label constraints for feature fusion [J], Pattern Recognition (PR), 2020. (CCF B类)

[13] Yang Li, Shichao Kan, Wenming Cao, Zhihai He, Learned Model Composition With Critical Sample Look-Ahead for Semi-Supervised Learning on Small Sets of Labeled Samples [J], IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2020, 9(31)3444-3445. (CCF B类

[14] Yang Li, Shichao Kan, Zhihai He, Unsupervised deep metric learning with transformed attention consistency and contrastive clustering loss [C], European Conference on Computer Vision (ECCV), 2020, 141-157. (CCF B类

[15] Shichao Kan, Yigang Cen, Zhihai He, Zhi Zhang, Linna Zhang, Yanhong Wang, Supervised deep feature embedding with handcrafted feature [J], IEEE Transactions on Image Processing (TIP), 2019, 28 (12) : 5809-5823. (CCF A类

[16] Shichao Kan, Lihui Cen, Xinwei Zheng, Yigang Cen, Zhenmin Zhu, Hengyou Wang, A Supervised Learning to Index Model for Approximate K-nearest Neighbor Image Retrieval [J], Signal Processing: Image Communication (SPIC), 2019, 78:494-502. (CCF C类

[17] Shi-Chao Kan, Yi-Gang Cen, Yi Cen, Yan-Hong Wang, Viacheslav Voronin, Vladimir Mladenovic, Ming Zeng, SURF binarization and fast codebook construction for image retrieval [J], Journal of visual communication and image representation (JVCIR), 2017, 49:104-114. (CCF C类


讲授课程

算法分析与设计(48课时,春夏学期,本科生)

面向对象编程(C++,48课时,秋冬学期,本科生)

教育经历

[1]   2016.9-2021.6

北京交通大学  |  信号与信息处理  |  工学博士学位  |  博士研究生毕业

[2]   2019.9-2020.10

美国密苏里大学哥伦比亚分校  |  图像和视频处理  |  联合培养博士

[3]   2014.9-2016.6

北京交通大学  |  电子与通信工程  |  硕士学位  |  硕士研究生毕业

[4]   2010.9-2014.6

北京交通大学  |  计算机科学与技术  |  学士学位  |  本科(学士)

工作经历

[1]   2021.9-至今

中南大学  |  计算机学院  |  讲师