tensorflow - CNN-> LSTM network for videos -


i have x number of videos , each video has different number of frames, let's y(x). frames size same videos 224x224x3. passing each frame cnn , outputs feature vector of 1024. now, want pass lstm. lstm batch_size, time_steps , number_of_feature required. how should decide value ? have 2 configurations in mind not know how should proceed.

  1. should break 1024 32 x 32 define time_steps , number_of_features , batch_size number of frames

  2. should time_step should corresponding number of frames , number_of_feature should 1024 , batch_size (?)

so depends on problem trying solve.

action classification using videos?

if trying predict action/event video have use num_of_frames time_steps, , batch_size number of videos want process together.

per frame object classification ?

in case can split features 32x32 time_steps,


Comments

Popular posts from this blog

Is there a better way to structure post methods in Class Based Views -

reflection - How to access the object-members of an object declaration in kotlin -

php - Doctrine Query Builder Error on Join: [Syntax Error] line 0, col 87: Error: Expected Literal, got 'JOIN' -