search

ristea / aed-mae

Other
23 stars 2 forks source link

about preprocessing #14

Open KirinHao opened 1 month ago

KirinHao commented 1 month ago

hi, I've noticed that your processing on the avenue dataset is all about image time frames, but the original dataset consists of AVI videos. Your code does not reflect the transformation from video to frames. I have converted it using my own method, but I still couldn't successfully replicate the experiment. Could you tell me how you processed the video into frames? Or could you share your avenue dataset? Thanks a lot! image

KirinHao commented 1 month ago

Some of your code uses time frames directly during preprocessing, and I have issues with this part. image