search

chaoyuaw / pytorch-coviar

Compressed Video Action Recognition
https://www.cs.utexas.edu/~cywu/projects/coviar/
GNU Lesser General Public License v2.1
503 stars 127 forks source link

How to obtain the residual information? #35

Open lshiwjx opened 5 years ago

lshiwjx commented 5 years ago

According to the code that the residual information is obtained by subtracting the two frames, can it be directly extracted from the compressed video?

chaoyuaw commented 5 years ago

I think it's possible. The main reason for the current implementation is for simplicity. FFMpeg is a fairly large and complicated system and I found the current implementation simpler.