search

KoDohwan / VT-TWINS

Video-Text Representation Learning via Differentiable Weak Temporal Alignment (PyTorch implementation for the CVPR 2022 paper)
10 stars 2 forks source link

How could I find the S2DTW code #4

Closed sakura2233565548 closed 2 years ago

sakura2233565548 commented 2 years ago

Hi, Thanks for your excellent work! I can not find the S2DTW core code, only find the soft_dtw in your released file. Could you share the S2DTW code with me? Thanks a lot!

KoDohwan commented 2 years ago

Hi, the code you want is already in the soft_dtw, this repository. Line 351-375 of the soft_dtw is an implementation of S2DTW but the name is still the soft_dtw with negative_dot_product. I'm sorry for the confusion.

sakura2233565548 commented 2 years ago

Hi, the code you want is already in the soft_dtw, this repository. Line 351-375 of the soft_dtw is an implementation of S2DTW but the name is still the soft_dtw with negative_dot_product. I'm sorry for the confusion.

Thanks a lot! I have found this code~