Learning Cross-modal Contrastive Features for Video Domain Adaptation.
Donghyun KimYi-Hsuan TsaiBingbing ZhuangXiang YuStan SclaroffKate SaenkoManmohan ChandrakerPublished in: CoRR (2021)
Keyphrases
- domain adaptation
- cross modal
- prior knowledge
- learning process
- video sequences
- low level
- key frames
- learning algorithm
- feature set
- labeled data
- unsupervised learning
- supervised learning
- image features
- feature space
- multi modal
- semi supervised
- semi supervised learning
- video data
- training examples
- visual data
- similarity measure