MSQNet: Actor-agnostic Action Recognition with Multi-modal Query.
Anindya MondalSauradip NagJoaquin M. PradaXiatian ZhuAnjan DuttaPublished in: CoRR (2023)
Keyphrases
- multi modal
- action recognition
- video search
- human actions
- bag of words
- query processing
- computer vision
- activity recognition
- action classification
- human detection
- multi modality
- recognition of human actions
- audio visual
- human activities
- recognizing actions
- action detection
- semantic concepts
- relevance feedback
- keywords
- recognizing human actions
- human object interactions
- image annotation
- pose estimation
- high dimensional
- view invariant
- machine learning
- action recognition in videos
- uni modal
- visual information
- retrieval systems
- information retrieval