search

shuangshuangguo / caffe2pytorch-tsn

Transform the caffe model to pytorch model for Temporal Segment Network
73 stars 16 forks source link

optical flow and frame #23

Open xiaofengWang-CCNU opened 4 years ago

xiaofengWang-CCNU commented 4 years ago

I used ffempt flow and frame,and i can not reproduce the result,so i think it may be the data problem which i use, but i have only one GPU,so i can not run the caffe code,can you support me the optical flow and frame ? Think You!

shuangshuangguo commented 4 years ago

Sorry, the dataset is already lost.

xiaofengWang-CCNU commented 4 years ago

ok,Thank you!     ------------------ Original ------------------ From:  "guoshuangshuang"<notifications@github.com>; Date:  Wed, Jul 1, 2020 09:24 PM To:  "shuangshuangguo/caffe2pytorch-tsn"<caffe2pytorch-tsn@noreply.github.com>; Cc:  "xiaofengWang-CCNU"<xiaofengWang@mails.ccnu.edu.cn>; "Author"<author@noreply.github.com>; Subject:  Re: [shuangshuangguo/caffe2pytorch-tsn] optical flow and frame (#23)

 

Sorry, the dataset is already lost.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

shuangshuangguo commented 4 years ago

I follow the guide to extract the optical flow information. https://github.com/yjxiong/temporal-segment-networks