Analyze video action with a deep learning model based on space-time and context information.
Lê Công LuanTran Minh TrietPublished in: SoICT (2023)
Keyphrases
- space time
- deep learning
- human actions
- video sequences
- spatial and temporal
- spatio temporal
- input video
- unsupervised learning
- dynamic scenes
- unsupervised feature learning
- video representation
- machine learning
- mental models
- video data
- scan statistic
- human motion
- moving objects
- dynamic textures
- weakly supervised
- training set
- image sequences
- interesting events