Make A Long Image Short: Adaptive Token Length for Vision Transformers.
Yichen ZhuYuqin ZhuJie DuYi WangZhicai OuFeifei FengJian TangPublished in: CoRR (2021)
Keyphrases
- image data
- input image
- image content
- spatial information
- image features
- image classification
- image segmentation
- image collections
- template matching
- image analysis
- image retrieval
- visual perception
- multiscale
- test images
- edge detection
- image representation
- single image
- real time
- image regions
- image matching
- grey level
- adaptive threshold
- pixel values
- image synthesis
- region of interest
- low level image processing
- color vision
- lighting conditions
- similarity measure
- image database
- image structure
- image pixels
- motion estimation
- image enhancement
- segmentation method