ET-SORT: A deep learning-based method for mesoscale eddies tracking in two-dimensional time-varying flow field.

Yu Zhang Bo Qin Hao Du

Published in: ICDLT (2021)

Keyphrases

deep learning
pairwise
flow field
segmentation method
similarity measure
optical flow
unsupervised learning
computer vision
optical flow estimation
image segmentation
feature vectors
supervised learning
clustering method
motion analysis