虚拟现实与图形学、计算机视觉、模式识别
个人简介:
孔德慧,教授、博士生导师;多媒体与智能软件技术北京市重点实验室骨干成员;CCF高级会员,IEEE会员,中国人工智能学会会员。
教育简历:
1989.9-1992.3在北京航空航天大学制造工程系获硕士学位。
1992.9-1996.3在北京航空航天大学制造工程系获博士学位。
工作履历
1996.4-现在,北京工业大学讲师、副教授、教授。
2004.2-2004.5,澳大利亚Deakin大学访问研究。
2005.1-2005.6,澳大利亚Sydney大学,双语教学培训。
学术兼职
担任TIP、Cybernetics、电子学报、计算机学报等多个国内外期刊的审稿人。
课程教学
本科生教学:《模糊数学》、《离散数学》、《计算机图形学》
研究生教学:《计算机图形学》
科研项目:
1) 基于广义生成对抗模型的情境相关人体姿态估计,国家自然科学基金项目(主持),2022.1-2025.12
2) 语义关联的室内场景对象建模及功能理解,国家自然科学基金项目(主持),2018.1-2021.12
3) 基于计算摄影学的中国手语光场获取研究,国家自然科学基金(主持),2014.1-2017.12
4) 高精度非接触式体感数据采集仪研制,国家自然科学基金仪器专项(合作单位负责人),2013.1-2016.12.
5) 室内场景引导的高可靠性人类家居行为理解与分析,北京自然科学基金项目(主持),2020.1-2022.12
荣誉和获奖
1) 2002年入选北京市科技新星计划(A类);
2) 2006年北京市人才强教计划中青年学术骨干;
3) 2007年获教育部高等学校科技进步一等奖;
4) 2009年获中国公路学会科学技术一等奖;
5) 2020年获吴文俊人工智能科技进步一等奖;
主要论文论著:
[1] Dehui Kong, Xiliang Li, Shaofan Wang, Jinghua Li, Baocai Yin. Learning visual-and-semantic knowledge embedding for zero-shot image classification. Applied Intelligence. 53(2): 2250-2264 (2023)
[2] Caixia Liu, Dehui Kong, Shaofan Wang, Qianxing Li, Jinghua Li, Baocai Yin. Multi-scale latent feature-aware network for logical partition based 3D voxel reconstruction. Neurocomputing 533: 22-34 (2023)
[3] Yongpeng Wu, Dehui Kong, Shaofan Wang, Jinghua Li, Baocai Yin. HPGCN: Hierarchical poselet-guided graph convolutional network for 3D pose estimation. Neurocomputing, 487: 243-256 (2022)
[4] Caixia Liu, Dehui Kong, Shaofan Wang, Jinghua Li, Baocai Yin. A Spatial Relationship Preserving Adversarial Network for 3D Reconstruction from a Single Depth View. ACM Transactions on Multimedia Computing, Communications, and Applications. 18(4): 110:1-110:22 (2022)
[5] Bin Sun, Dehui Kong, Shaofan Wang, Jinghua Li, Baocai Yin, Xiaonan Luo. GAN for vision, KG for relation: A two-stage network for zero-shot action recognition. Pattern Recognition. 126: 108563 (2022)
[6] Bin Sun, Shaofan Wang, Dehui Kong, Lichun Wang, Baocai Yin. Real-Time Human Action Recognition Using Locally Aggregated Kinematic-Guided Skeletonlet and Supervised Hashing-by-Analysis Model. IEEE Transactions on Cybernetics. 52(6): 4837-4849 (2022)
[7] Liu caixia, Dehui Kong, Wang shaofan, Li jinghua, Yin baocai. DLGAN: Depth-Preserving Latent Generative Adversarial Network for 3D Reconstruction. IEEE Transactions on Multimedia. 23: 2843-2856 (2021)
[8] Bin Sun, Dehui Kong, Shaofan Wang, Lichun Wang, Baocai Yin. Joint Transferable Dictionary Learning and View Adaptation for Multi-view Human Action Recognition. ACM Transactions on Knowledge Discovery from Data. 15(2): 32:1-32:23 (2021)
[9] Yongpeng Wu, Dehui Kong, Shaofan Wang, Jinghua Li, and Baocai Yin. An unsupervised real-time framework of human pose tracking from range image sequences. IEEE Transactions on Multimedia. 22(8): 2177-2190 (2020)
[10] Honglin Liu, Dehui Kong, Shaofan Wang, Baocai Yin. Sparse Pose Regression via Componentwise Clustering Feature Point Representation. IEEE Transactions on Multimedia. 18(7): 1233-1244 (2016)