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
To do so, I need to find all frames corresponding to a certain scene. In the example.ipynb notebook In[18] you provided a useful filter that helps to filter out the scenes based on tags. However, it seems that it's impossible to use such a filter to get particular scene_id.
Could you please suggest the best way to incorporate such a filter to get all the frames corresponding to a sequence displaying a certain agent moving in a certain scene?
Thank you!
The text was updated successfully, but these errors were encountered:
Hi! One of the options would be to iterate over the raw scenes protos using scenes_generator to find the scene of the interest. Also note the pre-rendered image filenames are {scene_id}_{request_track_id}.npy. So you can grep the filenames to identify the subfolder your scene is located in. It might save some time on iterations.
Hi!
I'm trying to create a visualization tool to produce video sequences like the one you provided on the official website: https://research.yandex.com/shifts/static/images/illus/motion-prediction-video.mp4
To do so, I need to find all frames corresponding to a certain scene. In the example.ipynb notebook In[18] you provided a useful filter that helps to filter out the scenes based on tags. However, it seems that it's impossible to use such a filter to get particular
scene_id
.Could you please suggest the best way to incorporate such a filter to get all the frames corresponding to a sequence displaying a certain agent moving in a certain scene?
Thank you!
The text was updated successfully, but these errors were encountered: