基本信息

                  汪航 博士

          微电子学院 副教授

    

 

联系方式:hangwang@xjtu.edu.cn

教育和工作经历

2024-至今,西安交通大学,电信学部微电子学院,副教授

2020-2023,西安交通大学,电信学部微电子学院,助理教授

2014-2020年,西安交通大学,电子科学与技术,工学博士

2010-2014年,哈尔滨工业大学,电子科学与技术,工学学士

站点计数器

个人简介

汪航,工学博士,西安交通大学微电子学院副教授。主要从事人工智能算法、视频信号处理集成电路、智能计算架构等领域的相关研究,涉及集成电路、模式识别等学科及其交叉学科。主持自然科学基金青年项目、重点研发计划子课题等多项国家级课题,作为核心骨干参与多项国家级课题,包括“核高基”重大专项、重点研发计划课题、自然科学基金面上项目等。在TCAS-I、TVLSI、ISCAS等国际高水平学术刊物与会议上发表论文多篇,获得国家发明专利授权多项,担任TCSVT、ISCAS等国际期刊和会议审稿人。

论文发表

 

  1. Baoting Li, Danqing Zhang, Pengfei Zhao, Hang Wang, et al., DQ-STP: An Efficient Sparse On-device Training Processor based on Low-rank Decomposition and Quantization for DNN[J]. IEEE Transactions on Circuits and Systems - I Regular Papers (TCAS-I), 2024. Accepted. 

  2. Danqing Zhang, Baoting Li, Hang Wang*, et al., An Efficient Sparse-Aware Summation Optimization Strategy for DNN Accelerator[C]. IEEE International Symposium on Circuits and Systems (ISCAS), 2024. (* Corresponding author) . Accepted. 

  3. He Dai, Hang Wang, Xuchong Zhang, et al., Memory-Efficient Batch Normalization by One-Pass Computation for On-Device Training[J]. IEEE Transactions on Circuits and Systems II: Express Briefs (TCAS-II), 2024. Accepted. 

  4. Hang Wang, Zhenyu Ding,  Cheng Cheng, et al., MDC-Net: Multi-Domain Constrained Kernel Estimation Network for Blind Image Super Resolution[J]. Computer Vision and Image Understanding, 2024

  5. Cheng Cheng, Lin Song, Ruoyi Xue, Hang Wang, Hongbin Sun et al. Meta-Adapter: An Online Few-shot Learner for Vision-Language Model[C]. NeurIPS 2023. 

  6. Cheng Cheng, Hang Wang, Xiang Liao,  Gang Cheng, et al., Continuity Preservation Network for Infrared Video Colorization[J]. Computer Vision and Image Understanding, 2023. 

  7. 汪航, 李宝婷, 张旭翀等, 深度神经网络在线训练硬件加速器的数据量化综述[J]. 微电子学与计算机. 2023.04.

  8. Zeyu Hao*, Hang Wang*, Xuchong Zhang, et al. Efficient Multi-Stage Network with Pixel-Wise Degradation Prediction for Real-time Motion Deblurring[J]. Computer Vision and Image Understanding, 2023. (* Equal contribution) .

  9. Baoting Li, Hang Wang*, Fujie Luo, et al. ACBN: Approximate Calculated Batch Normalization for Efficient DNN On-device Training Processor[J]. IEEE Transactions on Very Large Scale Integration Systems (TVLSI), 2023. (* Corresponding author) 

  10. Fujun Bai, Song Wang, Xuerong Jia, Yixin Guo, Bing Yu, Hang Wanget al., A Low-Cost Reduced-Latency DRAM Architecture with Dynamic Reconfiguration of Row Decoder[J]. IEEE Transactions on Very Large Scale Integration Systems (TVLSI), 2022. 

  11. Hang Wang, Cheng Cheng, Xuchong Zhang, et al., Towards High-Quality Thermal Infrared Image Colorization via Attention-Based Hierarchical Network[J]. Neurocomputing, 2022.

  12. Xuchong Zhang, Yongli Zhao, Hang Wang, Han Zhai, et al., End-to-end Learning of Self-rectification and Self-supervised Disparity Prediction for Stereo Vision[J]. Neurocomputing, 2022.

  13. Hang Wang, Cheng Cheng, Zeyu Hao, et al., High Quality Far Infrared Image Colorization Based on Generative Adversarial Network [C]. IEEE Asia Pacific Conference on Circuits and Systems (APCCAS),2021.

  14. Baoting Li, Hang Wang, Xuchong Zhang, et al. Dynamic Dataflow Scheduling and Computation Mapping Techniques for Efficient Depthwise Separable Convolution Acceleration [J]. IEEE Transactions on Circuits and Systems - I Regular Papers (TCAS-I), 2021, 68(8), 3279-3292.

  15. Hang Wang, Xiang Li, Xuchong Zhang, et al. A 4K×2K@60fps Multifunctional Video Display Processor for High Perceptual Image Quality[J]. IEEE Transactions on Circuits and Systems - I Regular Papers (TCAS-I), 2020, 451-463.

  16. Hang Wang, Tiancheng Wang, Longjun Liu, Hongbin Sun and Nanning Zheng. Efficient compression-based line buffer design for image/video processing circuits[J]. IEEE Transactions on Very Large Scale Integration Systems (TVLSI), 2019, 27(10): 2423-2433.

  17. Hang Wang, Tiancheng Wang, Yunqi Mi, Hongbin Sun and Nanning Zheng. A hardware-efficient post-processing algorithm for motion compensated frame rate up-conversion[C]. IEEE International Symposium on Circuits and Systems (ISCAS), 2019: 2158-1525.

  18. Hang Wang, Hongbin Sun, Xuchong Zhang, Nanning Zheng, et al. A 4K×2K@60fps multi-format multi-function display processor for high perceptual quality[C]. IEEE Asia Pacific Conference on Circuits and Systems, 2018: 427-430.

专利申请

  1. 汪航、孙宏滨、廖翔、张旭翀,一种基于改进顶帽变换的红外小目标增强方法[P].中国.发明专利申请号:202311808586X 申请时间:2023.12.

  2. 汪航、孙宏滨、郝泽宇、张旭翀,一种基于逐像素退化预测网络的多级去运动模糊方法[P].中国.发明专利申请号:2023108143089 申请时间:2023.07.

  3. 汪航、孙宏滨、闵子秋、张旭翀,一种用于图像盲超分辨率增强网络的退化核提取方法[P].中国.发明专利申请号:202210467732.6 申请时间:2022.04.

  4. 汪航、孙宏滨、程成、张旭翀,基于注意力机制的多尺度神经网络红外图像彩色化方法[P].中国.发明专利申请号:2022101996692. 申请时间:2022.03.

  5. 汪航、孙宏滨、闵子秋、张旭翀郑南宁,基于高斯-拉普拉斯金字塔的无监督学习X光图像增强方法[P],中国.发明专利申请号:202110130545.4. 授权时间: 2023.05.

  6. 张旭翀、孙宏滨、戴赫、汪航、赵永利、郑南宁,一种基于联合上采样卷积神经网络的双目立体匹配方法[P],中国.  发明专利申请号: 202010108633X. 申请时间: 2020.02.

  7. 孙宏滨、张旭翀、汪航、罗文、陈秋伯、郑南宁,一种基于局部自适应增益因子的联合上采样方法[P],中国. 发明专利授权号:ZL2017105871657. 授权时间:2020.07.

  8. 孙宏滨、王天成、汪航、赵鹏飞、郑南宁,基于数据压缩与搜索范围预测的运动估计存储架构设计方法,中国.发明专利授权号:ZL2018108045583. 授权时间:2020.06.

  9. 孙宏滨,任杰,李宝婷,张旭翀,汪航,郑南宁,一种深度可分离卷积神经网络加速器架构[P],中国. 发明专利申请号:202010628683.0.申请时间:2020.07.

  10. 孙宏滨,米云琪,汪航,张旭翀,郑南宁, 一种基于相关滤波的视频帧间快速运动估计方法[P],中国. 发明专利申请号:202010136402.X.授权时间:2023.04.

  11. 孙宏滨、汪航、张旭翀、陈秋伯、陶维乾、陈世强、罗文、郑南宁,集成电路布图设计:高画质数字视频显示处理芯片HPQ200[P],中国. 布图登记号码: BS.175527423