VideoSense at TRECVID 2011: Semantic Indexing from Light Similarity Functions-based Domain Adaptation with Stacking.
Emilie MorvantStéphane AyacheAmaury HabrardMiriam RediClaudiu TanaseBernard MérialdoBahjat SafadiFranck ThollardNadia DerbasGeorges QuénotPublished in: TRECVID (2011)
Keyphrases
- semantic indexing
- domain adaptation
- similarity function
- multimedia information retrieval
- pairwise
- semi supervised
- similarity measure
- cross domain
- multiple sources
- distance function
- semantic description
- labeled data
- kernel methods
- video retrieval
- transfer learning
- image content
- target domain
- semi supervised learning
- semantic concepts
- low level features
- multimedia documents
- active learning
- training data
- low dimensional
- supervised learning
- bag of visual words
- data analysis