Make a Long Image Short: Adaptive Token Length for Vision Transformers.

Qiqi Zhou Yichen Zhu

Published in: ECML/PKDD (2) (2023)

Keyphrases

image data
multiscale
single image
input image
image classification
image content
image features
visual perception
image retrieval
edge detection
template matching
image analysis
computer vision
region of interest
lighting conditions
image set
image pixels
low level vision
test images
segmentation method
similarity measure
image segmentation
image processing
vector field
natural images
markov random field
high resolution
image structure
pixel values
image synthesis