研究方向
连续手语识别与手语生成
面向连续手语识别（CSLR）与手语生成任务，研究跨模态语义对齐、时空特征建模、上下文建模与轻量化网络结构，成果发表于 CVPR、AAAI、ICASSP、Pattern Recognition、Neural Networks 等会议与期刊。
2026 年
- 2026DESign: Dynamic Context-Aware Convolution and Efficient Subnet Regularization for Continuous Sign Language Recognition, Sheng Liu, Yiheng Yu, Yuan Feng, Min Xu, Zhelun Jin, Yining Jiang, Tiantian Yuan, IEEE TCSVTCCF-BJCR Q1
[检索][BibTeX] - 2026SignMamba: Sparse Spatial–Temporal State-Space Modeling for Continuous Sign Language Recognition, Xu-Hua Yang, Gang Wang, Hong-Xiang Hu, Dong Wei, Zhi-Wei Chen, Sheng Liu, Yuan Feng, Neural Networks, 109530CCF-B
[链接][BibTeX] - 2026Focal-General Diffusion Model with Semantic Consistent Guidance for Sign Language Production, Yiheng Yu, Sheng Liu, Yuan Feng, Zhelun Jin, Yining Jiang, Min Xu, CVPR, 35915--35925CCF-A
[检索][BibTeX] - 2026SRPro: Semantic-aware Routing Attention with Prototype Anchored Alignment for Sign Language Production, Yiheng Yu, Sheng Liu, Yuan Feng, Zhelun Jin, Yining Jiang, Tiantian Yuan, Min Xu, ACM MMCCF-A
[检索][BibTeX] - 2026Improving Continuous Sign Language Recognition via Lightweight Adaptive Temporal Mixing, Yining Jiang, Sheng Liu, Yuan Feng, Min Xu, Yiheng Yu, ICASSP, 10462--10466CCF-B
[链接][BibTeX] - 2026SAKA: Spatially-Adaptive and Keyframe-Anchored Graph Network for Continuous Sign Language Recognition, Weiyi Ye, Xu-Hua Yang, Gang-Feng Ma, Dong Wei, Sheng Liu, Yuan Feng, ICASSP, 9907--9911CCF-B
[链接][BibTeX]
2025 年
- 2025OLMD: Orientation-aware Long-term Motion Decoupling for Continuous Sign Language Recognition, Yiheng Yu, Sheng Liu, Yuan Feng, Min Xu, Zhelun Jin, Xuhua Yang, AAAI, 9707-9715CCF-A
[链接][BibTeX] - 2025Fine-grained cross-modality consistency mining for Continuous Sign Language Recognition, Zhenghao Ke, Sheng Liu, Yuan Feng, Pattern Recognition Letters, 191: 23-30CCF-C
[DOI][BibTeX] - 2025Hierarchical spatial-temporal enhancement network for continuous sign language recognition, Min Xu, Sheng Liu, Yuan Feng, Yiheng Yu, Zhelun Jin, Xuhua Yang, ICASSP, 1--5CCF-B
[链接][BibTeX] - 2025Improving continuous sign language recognition via cross-frame interactions in expanded contextual spaces, Yiheng Yu, Sheng Liu, Yuan Feng, Min Xu, Zhelun Jin, Xuhua Yang, ICASSP, 1--5CCF-B
[链接][BibTeX] - 2025Design: Dynamic context-aware convolution and efficient subnet regularization for continuous sign language recognition, Sheng Liu, Yiheng Yu, Yuan Feng, Min Xu, Zhelun Jin, Yining Jiang, Tiantian Yuan, arXiv preprint arXiv:2507.03339
[链接][BibTeX] - 2025Cross-Modal Adaptive Prototype Learning for Continuous Sign Language Recognition, Dong Wei, Xu-Hua Yang, Yiyang Weng, Xuanyu Lin, Hongxiang Hu, Sheng Liu, IEEE TCSVT, 35(8): 7354-7367CCF-B
[DOI][BibTeX]
2024 年
- 2024DFCNet plus : Cross-modal dynamic feature contrast net for continuous sign language recognition, Yuan Feng, Nuoyi Chen, Yumeng Wu, Caoyu Jiang, Sheng Liu, Shengyong Chen, Image and Vision Computing, 151: 105260CCF-C
[DOI][BibTeX] - 2024Consistency-Driven Cross-Modality Transferring for Continuous Sign Language Recognition, Zhenghao Ke, Sheng Liu, Chengyuan Ke, Yuan Feng, SMC, 629--634CCF-C
[链接][BibTeX] - 2024CROSS-MODALITY CONSISTENCY MINING FOR CONTINUOUS SIGN LANGUAGE RECOGNITION WITH TEXT-DOMAIN EQUIVALENTS, Zhenghao Ke, Sheng Liu, Chengyuan Ke, Yuan Feng, Shengyong Chen, ICMECCF-B
[DOI][BibTeX] - 2024DPS-Net: Dual-Path Stimulation Network for Continuous Sign Language Recognition, Xiaopeng Tian, Sheng Liu, Yuan Feng, Jiantao Yang, Yineng Zhang, Songqi Pan, IJCNNCCF-C
[DOI][BibTeX]
人体运动预测与动作识别
研究人体骨骼序列建模、长期运动预测、三维人体网格重建与基于骨架的动作识别，提出高效的 MLP 与图卷积架构，成果发表于 IEEE TNNLS、ICASSP、IJCNN 等。
2026 年
- 2026DualSTFormer: A Dual-Stream Spatio-Temporal Transformer-GCN for 3D Human Pose Estimation, Sheng Liu, Haoyang Wang, Yuan Feng, Jianan Sun, Yining Jiang, Pengkai Tang, PRCVCCF-C
[检索][BibTeX]
2025 年
- 2025Highly Condensed All-MLP Architecture for Long-Term Human Motion Prediction, Sheng Liu, Shaobo Zhang, Fei Gao, Yuan Feng, IEEE Transactions on Neural Networks and Learning Systems, 36(10): 18274-18286ZJUT100/CCF-B中科院1区TOPJCR Q1
[DOI][BibTeX] - 2025Selective directed graph convolutional network for skeleton-based action recognition, Chengyuan Ke, Sheng Liu, Yuan Feng, Shengyong Chen, Pattern Recognition Letters, 190: 141-146CCF-C
[DOI][BibTeX] - 2025Cross-Scale Spatial Refinement Graph Convolutional Network for Skeleton-Based Action Recognition, Chengyuan Ke, Sheng Liu, Zhenghao Ke, Yuan Feng, Shengyong Chen, International Journal of Computational Intelligence Systems, 18(1): 76中科院4区
[DOI][PDF][BibTeX] - 2025Robust human motion prediction via integration of spatial and temporal cues, Shaobo Zhang, Sheng Liu, Fei Gao, Yuan Feng, Optoelectronics Letters, 21(8): 499--506中科院4区
[链接][BibTeX]
2024 年
- 2024Fp-gcn: A novel feature pyramid graph convolutional network for skeleton-based action recognition, Chengyuan Ke, Sheng Liu, Zhenghao Ke, Yuan Feng, SMC, 3715--3720CCF-C
[链接][BibTeX] - 2024POSE-HMR: HEURISTIC TRANSFORMER WITH POSTURAL PRIOR CONSTRAINTS FOR 3D HUMAN MESH RECONSTRUCTION, Songqi Pan, Sheng Liu, Yuan Feng, Yineng Zhang, Xiaopeng Tian, Jiantao Yang, ICASSP, 2835-2839CCF-B
[DOI][BibTeX] - 2024ATCE: Adaptive temporal context exploitation for weakly-supervised temporal action localization, Jiantao Yang, Sheng Liu, Yuan Feng, Xiaopeng Tian, Yineng Zhang, Songqi Pan, IJCNN, 1--8CCF-C
[链接][BibTeX] - 2024DYNAMIC MUTUAL-ACTIVATED TRANSFORMER FOR HUMAN MOTION PREDICTION, Shaobo Zhang, Sheng Liu, Fei Gao, Yuan Feng, ICASSP, 3040-3044CCF-B
[DOI][BibTeX] - 2024HCMLP: A Highly Condensed All-MLP Architecture for Extended Long-term Human Motion Prediction, Shaobo Zhang, Sheng Liu, Fei Gao, Yuan Feng, SMC, 4087--4092CCF-C
[链接][BibTeX]
2022 年
- 2022Human Interaction Recognition with Skeletal Attention and Shift Graph Convolution, Jin Zhou, Zhenhua Wang, Jiajun Meng, Sheng Liu, Jianhua Zhang, Shengyong Chen, IJCNNCCF-C
[DOI][BibTeX] - 2022PA-AWCNN: Two-stream Parallel Attention Adaptive Weight Network for RGB-D Action Recognition, Lu Yao, Sheng Liu, Chaonan Li, Siyu Zou, Shengyong Chen, Diyi Guan, ICRA, 8741-8747CCF-B
[DOI][BibTeX] - 2022Augmented Graph Attention with Temporal Gradation and Reorganization for Human Motion Prediction, Shaobo Zhang, Sheng Liu, Fei Gao, Shengyong Chen, ICIRA, 294-306
[DOI][BibTeX] - 2022HMD-former: a Transformer-based Human Mesh Deformer with Inter-layer Semantic Consistency, Siyu Zou, Sheng Liu, Chaonan Li, Lu Yao, Shengyong Chen, ICRA, 6409-6415CCF-B
[DOI][BibTeX]
复杂场景下事件目标检测、跟踪与图像恢复
研究复杂场景下退化视频中的目标检测、跟踪、图像盲去模糊、去雨去雾与小样本目标检测等恶劣环境下视觉任务，成果发表于 IEEE TIE、ACM TOMM 等期刊。
2024 年
- 2024Asymmetric Dual-Decoder U-Net for Joint Rain and Haze Removal, Yuan Feng, Yaojun Hu, Pengfei Fang, Sheng Liu, Yanhong Yang, Shengyong Chen, ACM Transactions on Multimedia Computing, Communications and Applications, 20(3): 87CCF-B
[DOI][PDF][BibTeX] - 2024SE-FewDet: Semantic-Enhanced Feature Generation and Prediction Refinement for Few-Shot Object Detection, Yineng Zhang, Sheng Liu, Yuan Feng, Songqi Pan, Xiaopeng Tian, Jiantao Yang, IJCNNCCF-C
[DOI][BibTeX] - 2024A Two-Stage Method for Aerial Tracking in Adverse Weather Conditions, Yuan Feng, Xinnan Xu, Nuoyi Chen, Quanjian Song, Lufang Zhang, Mathematics, 12(8): 1216中科院4区
[链接][BibTeX]
2019 年
- 2019L0 Sparse Regularization-Based Image Blind Deblurring Approach for Solid Waste Image Restoration, Sheng Liu, Yuan Feng, Shaobo Zhang, Hongzhang Song, Shengyong Chen, IEEE Transactions on Industrial Electronics, 66(12): 9837-9845ZJUT100中科院1区TOPJCR Q1
[DOI][BibTeX] - 2019Simultaneous 3D Motion Detection, Long-Term Tracking and Model Reconstruction for Multi-Objects, Sheng Liu, Yangqing Wang, Fengji Dai, Jingxiang Yu, International Journal of Humanoid Robotics, 16(4): 1950017中科院4区
[DOI][BibTeX]
具身智能三维视觉与图像分析
研究端到端具身智能导航、 RGB-D 物体分割、跨视场位姿估计、高光谱图像恢复与融合、零样本异常检测等问题，成果发表于 IEEE JSTARS、IEEE GRSL 等期刊。
2026 年
- 2026Embodied Safe Navigation via Per-Point Conformal Safety Allocation, Sheng Liu, Pengkai Tang, Yuan Feng, Haoyang Wang, Yining Jiang, Jianan Sun, PRCVCCF-C
[检索][BibTeX] - 2026Faclip: Learnable Fine-Grained Prompts and Multi-Scale Fusion for Zero-Shot Anomaly Detection, Shixin Zhao, Sheng Liu, Yuan Feng, Yiheng Yu, Zhelun Jin, Min Xu, ICASSP, 10407--10411CCF-B
[链接][BibTeX]
2021 年
- 2021CT-UNet: Context-Transfer-UNet for Building Segmentation in Remote Sensing Images, Sheng Liu, Huanran Ye, Kun Jin, Haohao Cheng, Neural Processing Letters, 53(6): 4257-4277CCF-C中科院4区JCR Q3
[DOI][BibTeX] - 2021Regularizing subspace representation for fusing hyperspectral and multispectral images, Yanhong Yang, Congcong Wang, Yuan Feng, Jianhua Zhang, Yuhui Zheng, Shengyong Chen, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 14: 12273--12286中科院2区
[链接][BibTeX] - 2021CT-UNet: An Improved Neural Network Based on U-Net for Building Segmentation in Remote Sensing Images, Huanran Ye, Sheng Liu, Kun Jin, Haohao Cheng, ICPR, 166-172CCF-C
[DOI][BibTeX]
数值计算方法
研究二维 quad-curl 问题的有限元方法、任意阶边值问题的再生核配置法等数值计算基础问题。