Replies: 3 comments 1 reply
-
Yes, the conversion could be done by |
Beta Was this translation helpful? Give feedback.
0 replies
-
Thanks a bunch for the help. I'll give this a try... I also noticed that fmtconv is used in parts of the vsmlrt scripts so I can give that a try. Do you have any full example script to convert files in real-time or other tips? |
Beta Was this translation helpful? Give feedback.
0 replies
-
import vapoursynth as vs
from vapoursynth import core
import vsmlrt
from vsmlrt import RIFEModel, Backend
backend = Backend.TRT(fp16=True, num_streams=2, use_cuda_graph=True, output_format=1)
src = core.lsmas.LWLibavSource("source.mp4")
ph = (src.height + 31) // 32 * 32 - src.height
pw = (src.width + 31) // 32 * 32 - src.width
padded = src.std.AddBorders(right=pw, bottom=ph).resize.Bicubic(format=vs.RGBH, matrix_in_s="709")
flt = vsmlrt.RIFE(padded, model=RIFEModel.v4_4, backend=backend)
res = flt.resize.Bicubic(format=vs.YUV420P8, matrix_s="709").std.Crop(right=pw, bottom=ph)
res.set_output() |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I'm trying to write a simple script to interpolate frames using RIFE. I'm completely new to VapourSynth (and vs-mlrt) so I apologize for the basic question.
I'm using AviSynthFilter (v1.4.6), MPC-HC, and vsml-rt v13.
Here is my script:
The error displayed is "vsmlrt.RIFE: only constant format 16/32 bit float input supported". VapourSynth reports the clip is using integer sample type with 8 bits per sample so it sounds like I need to convert it to FP16/FP32 before passing to RIFE?
Thanks for any help.
Beta Was this translation helpful? Give feedback.
All reactions