pytorch-pwc
pytorch-pwc copied to clipboard
sniklaus
how to import frames of river videos into PWC-Net codes
hi according to your python codes, in first UAV videos must extract to frames and Frames import to codes. Now i have 9000 Frames that extracted from mavic air combo videos (from Iran- Gorgan city Rivers). please Note to below section of code: arguments_strOne = './images/one.png' arguments_strTwo = './images/two.png' do i can import 9000 frames manually into above code? arguments_strOne = './images/frameone.png' arguments_strTwo = './images/frametwo.png' arguments_strOne = './images/framethree.png' arguments_strTwo = './images/framefour.png'
until last frame? i.e i must import two consecutive frames into above section code manually? it is time-consuming. i am confused. please guide me
What stops you from modifying the following part in the run.py to iterate over all the frames and estimate the flow between adjacent ones?
https://github.com/sniklaus/pytorch-pwc/blob/ed3af4f57d2409df97f74f012bd26a381ee454ed/run.py#L311-L324
If that is too slow you can batch multiple inputs but that requires more coding.
Hi thanks I think I should put two consecutive frames into below lines: arguments-strone = /data/frameone arguments-strtwo = /data/frametwo and I should modify this section of codes،🤔
در تاریخ پنجشنبه ۲۷ ژوئیهٔ ۲۰۲۳، ۱۸:۵۶ Simon Niklaus < @.***> نوشت:
What stops you from modifying the following part in the run.py to iterate over all the frames and estimate the flow between adjacent ones?
https://github.com/sniklaus/pytorch-pwc/blob/ed3af4f57d2409df97f74f012bd26a381ee454ed/run.py#L311-L324
If that is too slow you can batch multiple inputs but that requires more coding.
— Reply to this email directly, view it on GitHub https://github.com/sniklaus/pytorch-pwc/issues/64#issuecomment-1653848957, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGBWDOHOONOZ6LULPLQFU53XSKCC7ANCNFSM6AAAAAA2Z6FP6Q . You are receiving this because you authored the thread.Message ID: @.***>
Hi I have mistake excuse me Lines of code are : arguments_strone = /images/one.png arguments_strtwo = /images/two.png frame1 replaces one.png and frame2 replaces two.png I should modify this lines of codes (26 ,27) Yes?
در تاریخ جمعه ۲۸ ژوئیهٔ ۲۰۲۳، ۱۶:۰۱ hosein alhosseiny @.***> نوشت:
Hi thanks I think I should put two consecutive frames into below lines: arguments-strone = /data/frameone arguments-strtwo = /data/frametwo and I should modify this section of codes،🤔
در تاریخ پنجشنبه ۲۷ ژوئیهٔ ۲۰۲۳، ۱۸:۵۶ Simon Niklaus < @.***> نوشت:
What stops you from modifying the following part in the run.py to iterate over all the frames and estimate the flow between adjacent ones?
https://github.com/sniklaus/pytorch-pwc/blob/ed3af4f57d2409df97f74f012bd26a381ee454ed/run.py#L311-L324
If that is too slow you can batch multiple inputs but that requires more coding.
— Reply to this email directly, view it on GitHub https://github.com/sniklaus/pytorch-pwc/issues/64#issuecomment-1653848957, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGBWDOHOONOZ6LULPLQFU53XSKCC7ANCNFSM6AAAAAA2Z6FP6Q . You are receiving this because you authored the thread.Message ID: @.***>
hi there is error as "GetoptError: option -f not recognized"
in below code line:
for strOption, strArgument in getopt.getopt(sys.argv[1:], '', [ strParameter[2:] + '=' for strParameter in sys.argv[1::2] ])[0]:
i read that getopt is deprecated and argparse must be used instead.
i changed getopt to argparse but i have problems with argparse . i read about argparse very much but i am confused now.
please help me for fix GetoptError
hi i solved my error but now your scripts dont show any output in google colab. of course there is not any error. i am confused why dont show any output?
hi i defined a for loop in codes as following:
but the codes dont estimate flow between adjacent frames.
please help me.
My apologies, but I am unfortunately unable to provide such a level of support. Maybe try StackOverflow? Sorry!