1.个人简介
任子良,男,博士研究生,副教授,硕士研究生导师,CCF高级会员,2017年6月毕业于华南理工大学,先后在广州三星通讯研究院、中国科学院深圳先进技术研究院从事技术研发与科学研究等工作。目前主要从事机器视觉、动作识别与预测、行为理解和目标检测等方向的研究,先后在国内外学术期刊和国际学术会议发表学术论文50余篇,申请和授权国家发明专利50余项,主持广东省自然科学基金项目2项、东莞市科技特派员项目1项,作为核心成员/技术骨干参与科技部重点研发计划、国家自然科学基金、省市科学基金以及企业课题多项,获深圳市科学技术奖(技术发明奖)二等奖1项。
2.E-mail:renzl@dgut.edu.cn
3.工作经历
(1)2024/01-至今,东莞理工学院,计算机科学与技术学院,副教授
(2)2021/12-2023/12,东莞理工学院,计算机科学与技术学院,讲师
(3)2018/06-2021/11,中国科学院深圳先进技术研究院,博士后&助理研究员
(4)2012/07-2013/06,广州三星通讯研究院,工程师
4.科研项目
[1] 粤莞联合基金-地区培育项目,基于人机协作关系表示的行为意图理解方法研究,2023年1月-2025年12月,30万,主持;
[2] 广东省自然科学基金-面上项目,基于多模态特征关系嵌入表示的动作序列识别与理解方法研究,2023年1月-2025年12月,10万,主持;
[3] 东莞市科技特派员项目,智慧园区复杂场景中目标检测与异常行为分析及预警系统研究,2022年9月-2023年8月,10万,主持;
[4] 企业委托,AI行为分析系统,2024年1月-2024年12月,15万,主持。
5.学术论文
[1] Ronggui Liu, Ziliang Ren*, Wenhong Wei, Ziyang Zheng, Qieshi Zhang. Siamese Multi-View Masked Autoencoders for Skeleton-based Action Representation Learning, ICSMD, 2024.
[2] Huaigang Yang, Qieshi Zhang, Ziliang Ren∗, Huaqiang Yuan*, and Fuyong Zhang. Contrastive Learning with Cross-Part Bidirectional Distillation for Self-supervised Skeleton-based Action Recognition, Human-centric Computing and Information Sciences, 2024.
[3] Jiaqi Chen#, Ziliang Ren#, Qieshi Zhang*, Fuyong Zhang*, Wenguo Liu. Exploring the Potential of Residual Mechanism in Spiking Neural Networks for Human Action Recognition, Science China-Technological Sciences, 2024.
[4] Jiaqi Chen, Ziliang Ren∗, Wenhong Wei, Qieshi Zhang, and Xiangyang Gao. Incorporating Recursive and Stateful Self-Connection Learning of SNNs for Improved DVS Event Stream Processing, ICONIP, 2024.
[5] Ziyang Zheng, Ziliang Ren∗, Zhanhao Liang, Gulin Wang, Qieshi Zhang. MSGAT: multi-stage graph attention network for human motion prediction, IEEE International Conference on Image Processing (ICIP), 2024.
[6] Ziliang Ren, Qieshi Zhang, Qin Cheng, Zhenyu Xu, Shuai Yuan, Delin Luo*. Segment differential aggregation representation and supervised compensation learning of ConvNets for human action recognition, Science China-Technological Sciences, 67(1): 197-208, 2024.
[7] Miaomiao Jin, Ziliang Ren*, Wenhong Wei, Qian Chen, Ni An. Human Motion Prediction Based on Graph Convolutional Networks and Multilayer Perceptron, ICSMD, 2023.
[8] Huaigang Yang, Ziliang Ren*, Huaqiang Yuan, Qieshi Zhang, Wenhong Wei, Xiaoyu Hong. Multi-View Contrastive Self-supervised Triplet network for Skeleton-based Action Recognition, ICSMD, 2023, Best Student Paper.
[9] Huaigang Yang, Ziliang Ren*, Huaqiang Yuan, Zhenyu Xu and Jun Zhou. Contrastive self-supervised representation learning without negative samples for multimodal human action recognition, Frontiers in Neuroscience, (17)1225312: 1-14, 2023.
[10] Xiongjiang Xiao, Ziliang Ren*, Huan Li, Wenhong Wei, Zhiyong Yang, Huaide Yang. SlowFast Multimodality Compensation Fusion Swin Transformer Networks for RGB-D Action Recognition, Mathematics, 11(9): 2115, 2023.
[11] Li Luo, Ziliang Ren*, Yong Qin, Qieshi Zhang, Xiangyang Gao. Skeleton-Embedded Network for action recognition, IEEE International Conference on Real-time Computing and Robotics, 918-922, 2023.
[12] Jun Cheng, Ziliang Ren*, Qieshi Zhang, Xiangyang Gao, and Fusheng Hao. Cross-modality compensation convolutional neural networks for RGB-D action recognition, IEEE Transactions on Circuits and Systems for Video Technology, 32(3): 1498-1509, 2022.
[13] Ziliang Ren, Huaqiang Yuan, Wenhong Wei, Tiezhu Zhao, Qieshi Zhang*. Convolutional non-local spatial-temporal learning for multi-modality action recognition, Electronics Letters, 58(20): 765-767, 2022.
[14] Huaigang Yang, Ziliang Ren*, Huaqiang Yuan, Wenhong Wei, Qieshi Zhang and Zhaolong Zhang. Multi-scale and attention enhanced graph convolution network for skeleton-based violence action recognition, Frontiers in Neurorobotics, (16)1091361: 1-14, 2022.
[15] Xiongjiang Xiao, Ziliang Ren*, Wenhong Wei, Huan Li, Hua Tan. Shift Swin Transformer Multimodal Networks for Action Recognition in Videos, International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence, 1-6, 2022.
[16] Ziliang Ren, Qieshi Zhang, Jun Cheng*, Fusheng Hao, Xiangyang Gao. Segment spatial-temporal representation and cooperative learning of Convolution Neural Networks for multimodal-based action recognition, Neurocomputing, 433: 142-153, 2021.
[17] Ziliang Ren, Qieshi Zhang, Xiangyang Gao, Pengyi Hao, Jun Cheng*. Multi-modality Learning for Human Action Recognition, Multimedia Tools and Application, 80: 16185-16203, 2021.
[18] Ziliang Ren, Qieshi Zhang, Piye Qiao, Maolong Niu, Xiangyang Gao, and Jun Cheng*. Joint learning of convolution neural networks for RGB-D-based human action recognition, Electronics Letters, 56(21): 1112-1115, 2020.
6.专利授权和申请
[1] 任子良等,一种脉冲神经网络的训练方法,发明专利,CN202410888494.5,2024-07-04
[2] 任子良等,一种基于弱监督的自适应增强交互图卷积3D人体运动预测方法,发明专利,CN2024105109057,2024-04-26
[3] 任子良等,一种基于脉冲神经网络的人体动作识别方法,发明专利,CN202311680248.2,2023-12-08
[4] 任子良等,一种基于弱监督与长短期记忆网络的3D人体运动预测方法,发明专利,CN2023108 68509.7,2023-07-17
[5] 任子良等,基于知识蒸馏与多任务自监督学习的骨架行为识别方法,授权发明专利,ZL202310512443.8,2023-05-09
[6] 任子良等,融合堆叠LSTM 与SAC 算法的路径规划方法及系统,授权发明专利,ZL202310649008.X,2023-06-06
[7] 任子良等,一种定位方法、定位装置及存储介质,授权发明专利,ZL202310475470.2,2023-04-28
[8] 任子良等,动作序列识别和意图推断方法、装置、设备及存储介质, 授权发明专利,ZL202310335615.9,2023-04-23
[9] 任子良等,一种动作检测方法、装置、终端设备和存储介质,授权发明专利,ZL202110889116.5,2022-06-12
[10] 任子良等,一种行为识别方法、装置及终端设备,授权发明专利,ZL201910718037.0,2019-08-05
[11] 任子良等,一种人体动作识别和意图理解方法、终端设备及存储介质,授权发明专利,ZL202210675830.9,2024-01-10
[12] 任子良等,一种基于特征交互学习的动作识别方法及终端设备,授权发明专利,ZL202011078182.6,2020-10-10
7.获奖情况
[1] 2022年深圳市科学技术奖(技术发明奖)二等奖,人体动作识别与交互技术及应用,排名5/6。