Trailers12k: Improving Transfer Learning with a Dual Image and Video Transformer for Multi-label Movie Trailer Genre Classification.
Ricardo Montalvo-LezamaBerenice Montalvo-LezamaGibran Fuentes PinedaPublished in: CoRR (2022)
Keyphrases
- multi label
- transfer learning
- image classification
- learning tasks
- text categorization
- text classification
- genre classification
- image features
- labeled data
- machine learning
- image segmentation
- visual data
- image content
- image retrieval
- video content
- class labels
- video retrieval
- image annotation
- key frames
- neural network
- reinforcement learning
- video data
- active learning
- similarity measure
- unlabeled data
- semi supervised learning
- graph cuts
- expectation maximization
- low level
- training data
- computer vision
- video sequences
- music information retrieval