What Makes Training Multi-Modal Classification Networks Hard?
Weiyao WangDu TranMatt FeiszliPublished in: CVPR (2020)
Keyphrases
- multi modal
- training set
- training phase
- multi modality
- training samples
- training process
- high dimensional
- classification accuracy
- single modality
- text classification
- cross modal
- feature vectors
- audio visual
- image classification
- image annotation
- semantic concepts
- machine learning
- uni modal
- feature set
- supervised learning
- support vector machine
- feature space
- feature extraction
- feature selection
- video data
- object recognition