search

hx173149 / C3D-tensorflow

C3D is a modified version of BVLC tensorflow to support 3D ConvNets.
MIT License
589 stars 265 forks source link

I want to extract the characteristics of my video through tensorflow #89

Open 953585130 opened 5 years ago

953585130 commented 5 years ago

Hi， Thank you for your code! I built the environment and trained the C3D model from scratch, and the test also achieved good precision. Next I want to use this code to extract my own video features, add some other features (merge these features together), and classify them by SVM. So now I don't know how to get started. I didn't find a script with extracted features in the folder. Now I have encountered difficulties. Looking forward to your reply! thank you very much!

junaidwahid commented 5 years ago

@953585130 did you find the solution that how to extract features ?

953585130 commented 5 years ago

@junaidwahid Hi, I finally extracted the features through the caffe version of C3D. You can find the answer in this article "Learning Spatiotemporal Features with 3D Convolutional Networks". The code can be found on github.