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
Can you please tell me what the pickle file contains? I could understand that the dimensions are the same as image size and depth. I wasn't clear about a couple of things:
Do you convert RGB data to a different type?
Do you subsample frames from the video?
Are pkl files created individually for each Video file? If so, how are referenced in the code?
Will it be possible to share the code to convert video data to pkl?
The text was updated successfully, but these errors were encountered:
@jastir Hi. thank you for your interest in my work. my answers are as follows:
no its RGB.
for practical purpose, no. my method takes several consecutive frames and BLINDLY interpolates middle frames. for the UCF dataset, yes i need to skip some frames to serve as ground truth.
yes pkl is for each video file. For a common video, the code to preprocess is actually here. you then can just pickle it. but my code actually works for a folder of frames directly so you dont need to pickle it. For the UCF data, unfortunately ive lost my code recently due to a ransomware attack on my workplace's servers. if you still need it, please tell me. I will write it again.
Hi,
Can you please tell me what the pickle file contains? I could understand that the dimensions are the same as image size and depth. I wasn't clear about a couple of things:
Do you convert RGB data to a different type?
Do you subsample frames from the video?
Are pkl files created individually for each Video file? If so, how are referenced in the code?
Will it be possible to share the code to convert video data to pkl?
The text was updated successfully, but these errors were encountered: