Guoquan Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Guoquan Wang has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
Hong Liu
2 shared publications- Hong Liu
State Key Laboratory of General Artificial Intelligence, Peking University, Shenzhen Graduate School, Shenzhen, China
2 shared publications - Mengyuan Liu
Peking university, Shenzhen graduate school
2 shared publications - Wenhao Li
College of Computing and Data Science, Nanyang Technological University, Singapore
2 shared publications - Chao Feng
Shanghai Municipal Center for Disease Control and Prevention, Shanghai 200336, China
1 shared publication - Dasheng Lu
Division of Chemical Toxicity and Safety Assessment, Shanghai Municipal Center for Disease Control and Prevention, Shanghai 200336, China; State Environmental Protection Key Laboratory of Environmental Health Impact Assessment of Emerging Contaminants, Shanghai Municipal Center for Disease Control and Prevention, Shanghai 200336, China
1 shared publication
Latest Publications
- 21 April 2025
- Published by Elsevier in Pattern Recognition
- Vol. 167
Frequency-Aware Self-Supervised Group Activity Recognition with skeleton sequences
- 21 April 2025
- Published by Elsevier in Pattern Recognition
- Vol. 167
Frequency-Aware Self-Supervised Group Activity Recognition with skeleton sequences
- Open Access
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1672-1681
APP: Adaptive Pose Pooling for 3D Human Pose Estimation from Videos
Current advancements in 3D human pose estimation have attained notable success by converting 2D poses into their 3D counterparts. However, this approach is inherently influenced by the errors introduced by 2D pose detectors and overlooks the intrinsic spatial information embedded within RGB images. To address these challenges, we introduce a versatile module called Adaptive Pose Pooling (APP), which is compatible with many existing 2D-to-3D lifting models. The APP module includes three novel sub-modules: Pose-Aware Offsets Generation (PAOG), Pose-Aware Sampling (PAS), and Spatial Temporal Information Fusion (STIF). First, we extract latent features of the multi-frame lifting model. Then, a 2D pose detector is utilized to extract multi-level feature maps from the image. After that, PAOG generates offsets according to featuremaps. PAS uses offsets to sample featuremaps. Then, STIF can fuse PAS sampling features and latent features. This innovative design allows the APP module to simultaneously capture spatial and temporal information. We conduct comprehensive experiments on two widely used datasets: Human3.6M and MPI-INF-3DHP. Meanwhile, we employ various lifting models to demonstrate the efficacy of the APP module. Our results show that the proposed APP module consistently enhances the performance of lifting models, achieving state-of-the-art results. Significantly, our module achieves these performance boosts without necessitating alterations to the architecture of the lifting model. Our code is available at https://github.com/jinyanzhang/APP.
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1672-1681
Open AccessAPP: Adaptive Pose Pooling for 3D Human Pose Estimation from Videos
Current advancements in 3D human pose estimation have attained notable success by converting 2D poses into their 3D counterparts. However, this approach is inherently influenced by the errors introduced by 2D pose detectors and overlooks the intrinsic spatial information embedded within RGB images. To address these challenges, we introduce a versatile module called Adaptive Pose Pooling (APP), which is compatible with many existing 2D-to-3D lifting models. The APP module includes three novel sub-modules: Pose-Aware Offsets Generation (PAOG), Pose-Aware Sampling (PAS), and Spatial Temporal Information Fusion (STIF). First, we extract latent features of the multi-frame lifting model. Then, a 2D pose detector is utilized to extract multi-level feature maps from the image. After that, PAOG generates offsets according to featuremaps. PAS uses offsets to sample featuremaps. Then, STIF can fuse PAS sampling features and latent features. This innovative design allows the APP module to simultaneously capture spatial and temporal information. We conduct comprehensive experiments on two widely used datasets: Human3.6M and MPI-INF-3DHP. Meanwhile, we employ various lifting models to demonstrate the efficacy of the APP module. Our results show that the proposed APP module consistently enhances the performance of lifting models, achieving state-of-the-art results. Significantly, our module achieves these performance boosts without necessitating alterations to the architecture of the lifting model. Our code is available at https://github.com/jinyanzhang/APP.
- 9 February 2024
- Published by Elsevier in Pattern Recognition
- Vol. 150
Improving self-supervised action recognition from extremely augmented skeleton sequences
- 9 February 2024
- Published by Elsevier in Pattern Recognition
- Vol. 150
Improving self-supervised action recognition from extremely augmented skeleton sequences
- Open Access
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 8730-8739
Semantic-aware Consistency Network for Cloth-changing Person Re-Identification
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 8730-8739
Open AccessSemantic-aware Consistency Network for Cloth-changing Person Re-Identification
- 12 February 2022
- Published by Elsevier in Food Chemistry
- Vol. 383, 132431
Profiling of pesticides and pesticide transformation products in Chinese herbal teas
- 12 February 2022
- Published by Elsevier in Food Chemistry
- Vol. 383, 132431
Profiling of pesticides and pesticide transformation products in Chinese herbal teas
- 1 May 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 155-159
Mining and analysis of hot topics in Social Media
In social media, useful data information is mined to identify hot topics and predict the development direction of hot topics. The research method is to use web crawler to obtain relevant data, find new hot topics in media, get the original data, use Jieba to segment words, generate word segmentation results, obtain eigenvalues through Term Frequency - Inverse Document Frequency (TF-IDF) word frequency statistics, and use k-means algorithm to complete the clustering of eigenvalues, obtain the cluster results, and judge the hot topics in the cluster. In the experiment of advanced optimization hot topic, Linear Discriminant Analysis (LDA) method is used to classify the topic model, Gibbs sampling is used to sample and train the model, and K-means algorithm is used to achieve the purpose of clustering. The recall rate, accuracy rate and F value are calculated, and the experimental results of hot topics are summarized and analyzed to obtain the LDA model results of current hot topics in network media.
- 1 May 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 155-159
Mining and analysis of hot topics in Social Media
In social media, useful data information is mined to identify hot topics and predict the development direction of hot topics. The research method is to use web crawler to obtain relevant data, find new hot topics in media, get the original data, use Jieba to segment words, generate word segmentation results, obtain eigenvalues through Term Frequency - Inverse Document Frequency (TF-IDF) word frequency statistics, and use k-means algorithm to complete the clustering of eigenvalues, obtain the cluster results, and judge the hot topics in the cluster. In the experiment of advanced optimization hot topic, Linear Discriminant Analysis (LDA) method is used to classify the topic model, Gibbs sampling is used to sample and train the model, and K-means algorithm is used to achieve the purpose of clustering. The recall rate, accuracy rate and F value are calculated, and the experimental results of hot topics are summarized and analyzed to obtain the LDA model results of current hot topics in network media.