Paxion: Patching Action Knowledge in Video-Language Foundation Models.
Zhenhailong WangAnsel BlumeSha LiGenglin LiuJaemin ChoZineng TangMohit BansalHeng JiPublished in: CoRR (2023)
Keyphrases
- expert systems
- prior knowledge
- domain knowledge
- knowledge base
- multimedia
- video sequences
- human actions
- knowledge sharing
- statistical models
- knowledge acquisition
- neural network
- cognitive models
- video data
- probabilistic model
- real time
- knowledge management
- programming language
- action recognition
- spatio temporal
- conceptual model
- video content
- natural language
- video surveillance
- linguistic knowledge
- external world
- causal knowledge