ViT-AE++: Improving Vision Transformer Autoencoder for Self-supervised Medical Image Representations.
Chinmay PrabhakarHongwei LiJiancheng YangSuprosanna ShitBenedikt WiestlerBjoern H. MenzePublished in: MIDL (2023)
Keyphrases
- image representation
- multiscale
- image classification
- image content
- object recognition
- representation scheme
- bag of words
- feature representations
- visual words
- computer vision
- vision system
- sparse coding
- scene classification
- fuzzy logic
- image features
- image classification and retrieval
- image retrieval
- hierarchical structure
- quadtree
- receptive fields
- gaussian mixture modeling
- region segmentation
- natural scenes
- low level features
- probabilistic model
- machine learning
- spatial pyramid
- fault diagnosis
- sparse representation