Selur's Little Message Board

Hello Selur,

   I found a resize problem under specific setting.

   The input movie has resolution 1920x1080
   Since I want apply RealESERGAN to speed up the resize I applied Filtering->Vapoursynth->Misc->Lower res. before resize = 2

After having applied all the reaming filter settings and save the job, it is displayed the following error

[Image: Error-Resize.jpg]

The problem is the following:
   1) the initial size is: 1920x1080
       2) after the crop of 130 up/down the size is: 1920x820
       3) then after the resize by 2.0, the size is 960x410
       4) Hybrid add a border of 2, so that the size becomes: 960x412
       5) RealESERGAN resize the video to 3840x1648
       6) then Hybrid resize the video to 1920x820
       7) the Hybrid remove 8 pixels from the border
       8) the final size is 1920x812 -> ERROR

   I think the the step 7 should be applied before the step 6, in this way the output size should be correct
       I hope that you will be able to fix this issue

Thanks,
Dan

P.S.
I attached the debug file

Think I fixed this a few days ago.
Using your steps, my dev version ends with 1920x820:

Code:
# Imports

import vapoursynth as vs

# getting Vapoursynth core

import ctypes

import os

import site

core = vs.core

# Adding torch dependencies to PATH

path = site.getsitepackages()[0]+'/torch_dependencies/bin/'

ctypes.windll.kernel32.SetDllDirectoryW(path)

path = path.replace('\\', '/')

os.environ["PATH"] = path + os.pathsep + os.environ["PATH"]

# Loading Plugins

core.std.LoadPlugin(path="F:/Hybrid/64bit/vsfilters/Support/fmtconv.dll")

core.std.LoadPlugin(path="F:/Hybrid/64bit/vsfilters/SourceFilter/DGDecNV/DGDecodeNV.dll")

# source: 'G:\TestClips&Co\In_America_(2002).mkv'

# current color space: YUV420P8, bit depth: 8, resolution: 1920x1080, fps: 23.976, color matrix: 709, yuv luminance scale: limited, scanorder: progressive

# Loading G:\TestClips&Co\In_America_(2002).mkv using DGSource

clip = core.dgdecodenv.DGSource("J:/tmp/mkv_21c058149c797c35b163bbc419842782_853323747.dgi")# 23.976 fps, scanorder: progressive

# Setting detected color matrix (709).

clip = core.std.SetFrameProps(clip, _Matrix=1)

# Setting color transfer info (709), when it is not set

clip = clip if not core.text.FrameProps(clip,'_Transfer') else core.std.SetFrameProps(clip, _Transfer=1)

# Setting color primaries info (BT.709), when it is not set

clip = clip if not core.text.FrameProps(clip,'_Primaries') else core.std.SetFrameProps(clip, _Primaries=1)

# Setting color range to TV (limited) range.

clip = core.std.SetFrameProp(clip=clip, prop="_ColorRange", intval=1)

# making sure frame rate is set to 23.976

clip = core.std.AssumeFPS(clip=clip, fpsnum=24000, fpsden=1001)

clip = core.std.SetFrameProp(clip=clip, prop="_FieldBased", intval=0) # progressive

clip = core.std.CropRel(clip=clip, left=0, right=0, top=130, bottom=130)# cropping to 1920x820

# adjusting resolution before resizing 

clip = core.fmtc.resample(clip=clip, w=960, h=410, kernel="lanczos", interlaced=False, interlacedd=False)# before YUV420P8 after YUV420P16

from vsrealesrgan import realesrgan as RealESRGAN

clip = core.std.AddBorders(clip=clip, left=0, right=0, top=0, bottom=2) # add borders to archive mod 4 (vsRealESRGAN) - 960x412

# adjusting color space from YUV420P16 to RGBS for vsRealESRGAN

clip = core.resize.Bicubic(clip=clip, format=vs.RGBS, matrix_in_s="709", range_s="limited")

# resizing using RealESRGAN

clip = RealESRGAN(clip=clip, model=5, device_index=0) # 3840x1648

# resizing 3840x1648 to 1920x820

clip = core.std.CropRel(clip=clip, left=0, right=0, top=0, bottom=8) # removing borders (vsRealESRGAN) -  3840x1640

# adjusting resizing

clip = core.fmtc.resample(clip=clip, w=1920, h=820, kernel="spline64", interlaced=False, interlacedd=False)

# adjusting output color from: RGBS to YUV420P8 for x264Model

clip = core.resize.Bicubic(clip=clip, format=vs.YUV420P8, matrix_s="709", range_s="limited", dither_type="error_diffusion")

# set output frame rate to 23.976fps (progressive)

clip = core.std.AssumeFPS(clip=clip, fpsnum=24000, fpsden=1001)

# Output

clip.set_output()

=> will send you a link to my current dev via pm.

Cu Selur

Hello Selur,

The problem has been fixed with the last dev version.
It would be nice to add the option to resize using the mod (2,4,8,16,32) in Filtering->Vapoursynth->Misc->Lower res. before resize so that the add/remove border will be not more necessary.

Thanks,
Dan

The add/remove borders might still be necessary, i.e. if you resize to mod2, but the resizer requires mod16.
Also, I would rather add and remove borders than resizing and by abiding by a specific mod change the aspect ratio. Only minimal change, but adding and removing borders is fast and should not change the aspect ratio.

Cu Selur

Ps.: I'll look at it, since it should be easy to implement. => correction, is unexpectedly complicated to do this while maintaining the aspect ratio (as good as possible). => after some thinking about it and some calculation, I think it's not really possible without further restrictions.

Dan64

Selur

Dan64

Selur