An unsupervised learning based DIC method (一种无监督DIC方法)
Pytorch implementation of an unsupervised learning based DIC method.
The code provides three prediction examples, the imgs are from a public dataset, a self-built dataset and DIC challenge 1.0.
The six kinds of weights for public dataset and the weight for self-built dataset are provided.
The self-builtdataset is provided for your own further work.
Cuda
Pytorch
spatial-correlation-sampler
cv2
flowiz
...
The weights are available in https://drive.google.com/drive/folders/1K91tkiWumn0HKvvKuPibqwBQiBOWyi-u?usp=drive_link
The dataset is available in https://1drv.ms/u/s!Ag105hW9KtLsn3C_nkBgyol-H8Px?e=aLlg80
the sample from puclic dataset
the sample from self-build dataset
the sample from DIC challenge1.0
sample 13
sample 14