You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi there :)
I am trying to replicate your results and am currently on the data prepare steps.
In the Generate flow data from rgb and depth videos section you mention that I must run the docker image.
Is there a way for me to do it without that image? (I am running in an environment which is really strict about loading foreign images).
Thanks in advance,
Amit
The text was updated successfully, but these errors were encountered:
I recommend you to use this repo to extract the optical flow: https://github.com/open-mmlab/denseflow
But make sure to run the denseflow command with these flags:
denseflow <input_video> -a=tvl1 -b=20 -s=1 -st=jpg -nw=512 -nh=512 -o=<output_folder>
This code is able to make the extraction on a single video at the time. Although in the documentation it appears to be able to extract the flows of all the videos from a directory, that didn't work for me. Instead I made it extract the flow on each video at a time. Make sure to do a for loop on the directory in which you have the data stored and call the denswflow command for each one of them.
Hi there :)
I am trying to replicate your results and am currently on the data prepare steps.
In the Generate flow data from rgb and depth videos section you mention that I must run the docker image.
Is there a way for me to do it without that image? (I am running in an environment which is really strict about loading foreign images).
Thanks in advance,
Amit
The text was updated successfully, but these errors were encountered: