ViLPAct: A Benchmark for Compositional Generalization on Multimodal Human Activities.
Terry Yue ZhuoYaqing LiaoYuecheng LeiLizhen QuGerard de MeloXiaojun ChangYazhou RenZenglin XuPublished in: CoRR (2022)
Keyphrases
- human activities
- activity recognition
- human activity recognition
- human computer interaction
- sensor data
- visual surveillance
- smart home
- human behaviour
- action recognition
- daily life
- ubiquitous computing
- human interactions
- human knowledge
- temporal structure
- data streams
- machine learning
- video images
- human actions
- information extraction
- data model
- computer vision
- learning algorithm