Step 3: Extract the zip to a folder with an easy location with no spaces. (e.g. SuperSloMo)
Step 4: Download the pre-trained model here:
Place in SuperSloMo directory.
Step 5: Download ffmpeg from here: https://ffmpeg.zeranoe.com/builds/
Choose correct options for system & stable.
Unzip and rename to a simple name "ffmpeg" and place in the same parent directory as others.
Step 6: Open an Anaconda prompt. (nur für Windows User. MacOS/Linux User, einfach Terminal starten und unten genannten Befehl ausführen )
Use this link to generate the correct command: https://pytorch.org/get-started/locally/
e.g. conda install pytorch torchvision cudatoolkit=9.0 -c pytorch
And paste into the Anaconda prompt.
Step 7: Make an "Input" folder and place the footage you want to increase the framerate of inside it. Also make an "Output" folder for the finished conversions.
Step 8: Use "cd" in Anaconda to change the path to the folder that contains "video_to_slomo.py" & then run the process with the following command and arguments:
Mac/Linux User:
cd Super-SloMo-master
und z.B. python video_to_slomo.py --ffmpeg ffmpeg\bin\ --video /Users/***/Super-SloMo-master/Input/BMPCCProRes.mov --sf 4 --checkpoint SuperSloMo.ckpt --fps 120 --output /Users/***/Super-SloMo-master/Output/BMPPCProResSloMo.mov --batch_size 1
(Prozess dauert aber sehr lange)
Change the locations above to match the paths on your computer.
sf = multiple of frame rate
fps = target frame rate
batch_size = how quickly it will run/how much of the system will be used. (1 is lowest)
Step 9: Import the finished video into video editing software and lower the speed by the corresponding amount. For sf 4, set speed to 25%, etc.
"Deine Zeit ist begrenzt. Verschwende sie nicht damit, das Leben eines Anderen zu leben"
(Steve Jobs)
Rechtliche Notiz: Wir übernehmen keine Verantwortung für den Inhalt der Beiträge
und behalten uns das Recht vor, Beiträge mit rechtswidrigem oder anstößigem Inhalt zu löschen.