Wang-Wang Yu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Wang-Wang Yu has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Yong-Jie Li
MOE Key Laboratory for Neuroinformation, School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, China
4 shared publications - Hongmei Yan
The Clinical Hospital of Chengdu Brain Science Institute, MOE Key Lab for Neuroinformation, School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, China
2 shared publications - Kai-Fu Yang
MOE Key Laboratory for Neuroinformation, School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, China
2 shared publications - Jingwen Jiang
West China School of Public Health and West China Fourth Hospital, State Key Laboratory of Biotherapy, Sichuan University Tianfu Jincheng Laboratory Chengdu China
1 shared publication Philippe Zhang
University of Western Brittany
1 shared publicationShiye Wang
Henan Kaifeng College of Science Technology and Communication
1 shared publication
Latest Publications
- 12 August 2025
- Published by Springer Nature
- p. 11-20
Classification and Segmentation of Intrapartum Ultrasound Images with Deep Learning Models
- 12 August 2025
- Published by Springer Nature
- p. 11-20
Classification and Segmentation of Intrapartum Ultrasound Images with Deep Learning Models
- 28 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 47 (8) , 6912-6928
Weakly Supervised Micro- and Macro-Expression Spotting Based on Multi-Level Consistency
Most micro- and macro-expression spotting methods in untrimmed videos suffer from the burden of video-wise collection and frame-wise annotation. Weakly supervised expression spotting (WES) based on video-level labels can potentially mitigate the complexity of frame-level annotation while achieving fine-grained frame-level spotting. However, we argue that existing weakly supervised methods are based on multiple instance learning (MIL) involving inter-modality, inter-sample, and inter-task gaps. The inter-sample gap is primarily from the sample distribution and duration. Therefore, we propose a novel and simple WES framework, MC-WES, using multi-consistency collaborative mechanisms that include modal-level saliency, video-level distribution, label-level duration and segment-level feature consistency strategies to implement fine frame-level spotting with only video-level labels to alleviate the above gaps and merge prior knowledge. The modal-level saliency consistency strategy focuses on capturing key correlations between raw images and optical flow. The video-level distribution consistency strategy utilizes the difference of sparsity in temporal distribution. The label-level duration consistency strategy exploits the difference in the duration of facial muscles. The segment-level feature consistency strategy emphasizes that features under the same labels maintain similarity. Experimental results on three challenging datasets–CAS(ME)$^{2}$ , CAS(ME)$^{3}$ , and SAMM-LV–demonstrate that MC-WES is comparable to state-of-the-art fully supervised methods.- 28 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 47 (8) , 6912-6928
Weakly Supervised Micro- and Macro-Expression Spotting Based on Multi-Level Consistency
Most micro- and macro-expression spotting methods in untrimmed videos suffer from the burden of video-wise collection and frame-wise annotation. Weakly supervised expression spotting (WES) based on video-level labels can potentially mitigate the complexity of frame-level annotation while achieving fine-grained frame-level spotting. However, we argue that existing weakly supervised methods are based on multiple instance learning (MIL) involving inter-modality, inter-sample, and inter-task gaps. The inter-sample gap is primarily from the sample distribution and duration. Therefore, we propose a novel and simple WES framework, MC-WES, using multi-consistency collaborative mechanisms that include modal-level saliency, video-level distribution, label-level duration and segment-level feature consistency strategies to implement fine frame-level spotting with only video-level labels to alleviate the above gaps and merge prior knowledge. The modal-level saliency consistency strategy focuses on capturing key correlations between raw images and optical flow. The video-level distribution consistency strategy utilizes the difference of sparsity in temporal distribution. The label-level duration consistency strategy exploits the difference in the duration of facial muscles. The segment-level feature consistency strategy emphasizes that features under the same labels maintain similarity. Experimental results on three challenging datasets–CAS(ME)$^{2}$ , CAS(ME)$^{3}$ , and SAMM-LV–demonstrate that MC-WES is comparable to state-of-the-art fully supervised methods.- 13 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Affective Computing
- Vol. 15 (1) , 223-240
LGSNet: A Two-Stream Network for Micro- and Macro-Expression Spotting With Background Modeling
Micro- and macro-expression spotting in an untrimmed video is a challenging task, due to the mass generation of false positive samples. Most existing methods localize higher response areas by extracting hand-crafted features or cropping specific regions from all or some key raw images. However, these methods either neglect the continuous temporal information or model the inherent human motion paradigms (background) as foreground. Consequently, we propose a novel two-stream network, named Local suppression and Global enhancement Spotting Network (LGSNet), which takes segment-level features from optical flow and videos as input. LGSNet adopts anchors to encode expression intervals and selects the encoded deviations as the object of optimization. Furthermore, we introduce a Temporal Multi-Receptive Field Feature Fusion Module (TMRF $^{3}$ M) and a Local Suppression and Global Enhancement Module (LSGEM), which help spot short intervals more precisely and suppress background information. To further highlight the differences between positive and negative samples, we set up a large number of random pseudo ground truth intervals (background clips) on some discarded sliding windows to accomplish background clips modeling to counteract the effect of non-expressive face and head movements. Experimental results show that our proposed network achieves state-of-the-art performance on the CAS(ME) $^{2}$ , CAS(ME) $^{3}$ and SAMM-LV datasets.
- 13 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Affective Computing
- Vol. 15 (1) , 223-240
LGSNet: A Two-Stream Network for Micro- and Macro-Expression Spotting With Background Modeling
Micro- and macro-expression spotting in an untrimmed video is a challenging task, due to the mass generation of false positive samples. Most existing methods localize higher response areas by extracting hand-crafted features or cropping specific regions from all or some key raw images. However, these methods either neglect the continuous temporal information or model the inherent human motion paradigms (background) as foreground. Consequently, we propose a novel two-stream network, named Local suppression and Global enhancement Spotting Network (LGSNet), which takes segment-level features from optical flow and videos as input. LGSNet adopts anchors to encode expression intervals and selects the encoded deviations as the object of optimization. Furthermore, we introduce a Temporal Multi-Receptive Field Feature Fusion Module (TMRF $^{3}$ M) and a Local Suppression and Global Enhancement Module (LSGEM), which help spot short intervals more precisely and suppress background information. To further highlight the differences between positive and negative samples, we set up a large number of random pseudo ground truth intervals (background clips) on some discarded sliding windows to accomplish background clips modeling to counteract the effect of non-expressive face and head movements. Experimental results show that our proposed network achieves state-of-the-art performance on the CAS(ME) $^{2}$ , CAS(ME) $^{3}$ and SAMM-LV datasets.
- 1 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2563-2570
Fast Face Recognition Model without Pruning
This paper proposes a real-time high accuracy face recognition model which can be deployed on performanceconstrained embedded platforms or mobile terminals without pruning. In the proposed model, we optimize the network structure by balancing the memory access cost (MAC) and Floatingpoint Operations (FLOPs) and propose a new factor to measure the efficiency of learning parameters. Based on these two metrics, we designed two structures called Simple SqueezeFaceNet (SSN) and Channel-Split Network (CSN) to achieve a good balance between the high precision and high FPS (Frames per Second). After being trained with the refined MS1M-refine-v2 dataset, our architectures achieve 0.992 (or 0.990) accuracy on the Labeled Faces in the Wild (LFW) with 155 (or 180) FPS. This indicates that our models can also be applied to real-time multi-face recognition in video.
- 1 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2563-2570
Fast Face Recognition Model without Pruning
This paper proposes a real-time high accuracy face recognition model which can be deployed on performanceconstrained embedded platforms or mobile terminals without pruning. In the proposed model, we optimize the network structure by balancing the memory access cost (MAC) and Floatingpoint Operations (FLOPs) and propose a new factor to measure the efficiency of learning parameters. Based on these two metrics, we designed two structures called Simple SqueezeFaceNet (SSN) and Channel-Split Network (CSN) to achieve a good balance between the high precision and high FPS (Frames per Second). After being trained with the refined MS1M-refine-v2 dataset, our architectures achieve 0.992 (or 0.990) accuracy on the Labeled Faces in the Wild (LFW) with 155 (or 180) FPS. This indicates that our models can also be applied to real-time multi-face recognition in video.