PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 23, 2026Computers, materials & continua/Computers, materials & continua (Print)Open Access

Group Activity Recognition in Crowded Scenes Using Multi-Stage Feature Optimization and ST-GCN-LSTM Networks

View Full Paper
Ask AI
Bookmark
Share

Authors

MAMohammed AlnusayriTXTingting XueSKSaleha Kamal

Discussion

Loading...

Member takes

Overview

This work demonstrates enhanced group activity recognition in crowded scenes, suggesting improvements for intelligent surveillance.

Key Points

  • This research aims to improve group activity recognition in public environments despite challenges like occlusions and dynamic formations.
  • Developed a multi-modal framework integrating silhouette and pose information for recognition.
  • Utilized YOLOv11 for detection, SOLOv2 for segmentation, and AlphaPose for skeleton extraction.
  • Employed a three-stage feature optimization process using K-PCA, mutual information ranking, and genetic algorithms.
  • Achieved 96.80% accuracy on the Collective Activity Dataset, outperforming existing methods.
  • Successfully captured collective behavior by integrating various feature extraction techniques.
  • Demonstrated scalability and adaptability for smart city applications.

Cite This Study

Alnusayri et al. (2026) studied this question.

synapsesocial.com/papers/69e9b9a285696592c86ec342https://doi.org/10.32604/cmc.2026.074115
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Skeleton-based Group Activity Recognition via Spatial-Temporal Panoramic Graph2024
  2. 2A multi-context fusion-aware graph modelling for group activity recognition using pose-conditioned spatial encoding and actor relations2026
  3. 3Skeleton-Based Action Recognition with Spatial-Structural Graph Convolution2024
  4. 4Intrinsic Topology and Multi-Scale Temporal Modeling for Skeleton-Based Human Action Recognition in Smart Surveillance Systems2026
  5. 5Deep Learning-Based Control System for Context-Aware Surveillance Using Skeleton Sequences from IP and Drone Camera video2025