search

JJBOY / BMN-Boundary-Matching-Network

A pytorch-version implementation codes of paper: "BMN: Boundary-Matching Network for Temporal Action Proposal Generation", which is accepted in ICCV 2019.
MIT License
291 stars 64 forks source link

about feature extraction #10

Closed sunnymoon155 closed 5 years ago

sunnymoon155 commented 5 years ago

Hi, thanks for the code. I'd like to try this on my own datasets, but when I was extracting features by using TSN, I can only get RGB.jpg and Flow_x_y.jpg. How can I get csv features which is used in your code? (like csv_mean_100/v_oOnKQgQZOZ0.csv) Or can I use jpg to do training and testing?

JJBOY commented 5 years ago

Please follow the TSN's guide. You only got RGB and Flow because you only used dense flow to the original videos. You should use TSN network to get features then.

semchan commented 4 years ago

Please follow the TSN's guide. You only got RGB and Flow because you only used dense flow to the original videos. You should use TSN network to get features then.

Thanks for your great work. If I want only use RGB features of the anet v1.3 which the author provided, is that I can only use the first 200 feature's in the .csv file? thanks a lot.