计算机视觉

主要开展目标杂乱、时序信号速度快、数据域多变、监督信号弱等复杂条件下的多媒体视觉计算研究，包括目标检测、异常检测、音视频解析、目标跟踪和动态估计等。探索域迁移与域泛化学习、类增量学习、在线学习、弱监督学习以及多目标优化等理论与方法，提出了特权知识蒸馏的弱监督学习、身份特征增强的统一追踪框架等方法，有效提高了目标与事件识别的准确度、检测定位的完整度、长期预测追踪的稳定性等。在器件缺陷检测、人群安全管控、异常事件预警等领域开展了技术应用。

代表性论文

[1] Sisi You, Hantao Yao, Bing-Kun Bao*, Changsheng Xu. Multi-object Tracking with Spatial-Temporal Tracklet Association. ACM Transactions on Multimedia Computing, Communications and Applications (ACM TOMM) 2024 【论文】

[2] Tianshan Liu, Kin-man Lam, Bing-kun Bao*. Label Text-aided Hierarchical Semantics Mining for Panoramic Activity Recognition. ACM Multimedia (ACM MM) 2024 【论文】

[3] Tianshan Liu, Kin-Man Lam, Bing-Kun Bao*. Injecting Text Clues for Improving Anomalous Event Detection From Weakly Labeled Videos. IEEE Transactions on Image Processing (TIP) 2024 【论文】

[4] Tianshan Liu, Kin-Man Lam, Bing-Kun Bao*. A Memory-Assisted Knowledge Transferring Framework with Curriculum Anticipation for Weakly Supervised Online Activity Detection. International Journal of Computer Vision (IJCV) 2024 【论文】

[5] Jiao, Yifan, Hantao Yao, Changsheng Xu. Dual Instance-consistent Network for Cross-domain Object Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) 2023 【论文】

[6] Sisi You, Hantao Yao, Bing-kun Bao, Changsheng Xu. UTM: A Unified Multiple Object Tracking Model with Identity-aware Feature Enhancement. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2023 【论文】

[7] Jie Fu, Junyu Gao, Bing-Kun Bao, Changsheng Xu. Multimodal Imbalance-Aware Gradient Modulation for Weakly-Supervised Audio-Visual Video Parsing. IEEE Transactions on Circuits and Systems for Video Technology (TCSVT) 2023 【论文】