A study of DQN using VisionTransformer as an image extractor.

Toshiki Hatano Toi Tsuneda Satoshi Yamane

Published in: GCCE (2021)

Keyphrases

multiscale
image features
image data
image retrieval
image analysis
single image
image content
input image
empirical studies
image classification
image pixels
image collections
image representation
keypoints
template matching
image matching
image segmentation
lighting conditions
edge detection
segmentation method
image noise
test images
vector field
structure from motion
segmentation algorithm
post processing
feature points
low level