Deoldify Vapoursynth filter

[Selur]

@Dan64: using:

# Imports import sys import os import vapoursynth as vs # getting Vapoursynth core core = vs.core # Limit frame cache to 48449MB core.max_cache_size = 48449 # Import scripts folder scriptPath = 'F:/Hybrid/64bit/vsscripts' sys.path.insert(0, os.path.abspath(scriptPath)) # loading plugins core.std.LoadPlugin(path="F:/Hybrid/64bit/vsfilters/MiscFilter/MiscFilters/MiscFilters.dll") core.std.LoadPlugin(path="F:/Hybrid/64bit/vsfilters/SourceFilter/BestSource/BestSource.dll") # Import scripts import vsdeoldify as havc import validate # Source: 'G:\TestClips&Co\files\test.avi' # Current color space: YUV420P8, bit depth: 8, resolution: 640x352, frame rate: 25fps, scanorder: progressive, yuv luminance scale: limited, matrix: 470bg, format: MPEG-4 Visual # Loading 'G:\TestClips&Co\files\test.avi' using BestSource clip = core.bs.VideoSource(source="G:/TestClips&Co/files/test.avi", cachepath="J:/tmp/test_bestSource", track=0, hwdevice="opencl") frame = clip.get_frame(0) # setting color matrix to 470bg. clip = core.std.SetFrameProps(clip, _Matrix=vs.MATRIX_BT470_BG) # setting color transfer (vs.TRANSFER_BT601), if it is not set. if validate.transferIsInvalid(clip): clip = core.std.SetFrameProps(clip=clip, _Transfer=vs.TRANSFER_BT601) # setting color primaries info (to vs.PRIMARIES_BT470_BG), if it is not set. if validate.primariesIsInvalid(clip): clip = core.std.SetFrameProps(clip=clip, _Primaries=vs.PRIMARIES_BT470_BG) # setting color range to TV (limited) range. prop_name = '_Range' if core.core_version.release_major >= 74 else '_ColorRange' clip = core.std.SetFrameProps(clip=clip, **{prop_name: vs.RANGE_LIMITED}) # making sure frame rate is set to 25fps clip = core.std.AssumeFPS(clip=clip, fpsnum=25, fpsden=1) # making sure the detected scan type is set (detected: progressive) clip = core.std.SetFrameProps(clip=clip, _FieldBased=vs.FIELD_PROGRESSIVE) # scan type: progressive original = clip # changing color matrix from '470bg' to '709' for vsHAVC clip = core.resize.Bicubic(clip, matrix_in_s="470bg", matrix_s="709") # adjusting color space from YUV420P8 to RGB24 for vsHAVC clip = core.resize.Bicubic(clip=clip, format=vs.RGB24, matrix_in_s="709", range_in_s="limited", range_s="full") # adding colors using HAVC clip = havc.HAVC_main(clip, EnableDeepEx=True, DeepExMethod=0, DeepExRefMerge=0, ScFrameDir=None, DeepExModel=0, DeepExEncMode=0, DeepExMaxMemFrames=0) # internally changing color matrix for YUV<>RGB to '470bg' undoing color matrix change for vsHAVC # adjusting output color from YUV420P8 to YUV420P10 for NVEncModel original = core.resize.Bicubic(clip=original, format=vs.YUV420P10) # adjusting output color from RGB24 to YUV420P10 for NVEncModel clip = core.resize.Bicubic(clip=clip, format=vs.YUV420P10, matrix_s="470bg", range_in_s="full", range_s="limited") # additional resize to match target color sampling original = core.text.Text(clip=original,text="Original",scale=1,alignment=7) clip = core.text.Text(clip=clip,text="Filtered",scale=1,alignment=7) stacked = core.std.StackHorizontal([original,clip]) # set output frame rate to 25fps (progressive) stacked = core.std.AssumeFPS(clip=stacked, fpsnum=25, fpsden=1) # output stacked.set_output()

the HAVC call is:

clip = havc.HAVC_main(clip, EnableDeepEx=True, DeepExMethod=0, DeepExRefMerge=0, ScFrameDir=None, DeepExModel=0, DeepExEncMode=0, DeepExMaxMemFrames=0)

I get:

2026-04-25 20:03:04.209 Failed to evaluate the script: Python exception: CMNET2: number of reference frames must be at least 2, found 2 Traceback (most recent call last): File "vapoursynth.pyx", line 3524, in vapoursynth._vpy_evaluate File "vapoursynth.pyx", line 3525, in vapoursynth._vpy_evaluate File "J:\tmp\tempPreviewVapoursynthFile20_01_43_472.vpy", line 44, in clip = havc.HAVC_main(clip, EnableDeepEx=True, DeepExMethod=0, DeepExRefMerge=0, ScFrameDir=None, DeepExModel=0, DeepExEncMode=0, DeepExMaxMemFrames=0) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "f:\hybrid\64bit\vapoursynth\Lib\site-packages\vsdeoldify\__init__.py", line 321, in HAVC_main return HAVC_main_presets(clip, Preset, FrameInterp, ColorModel, CombMethod, VideoTune, ColorFix, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "f:\hybrid\64bit\vapoursynth\Lib\site-packages\vsdeoldify\__init__.py", line 503, in HAVC_main_presets clip_colored = HAVC_main_colorizer(clip, Preset, ColorModel, CombMethod, VideoTune, ColorFix, ColorTemp, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "f:\hybrid\64bit\vapoursynth\Lib\site-packages\vsdeoldify\__init__.py", line 840, in HAVC_main_colorizer clip_colored = HAVC_deepex(clip=clip, clip_ref=clip_ref, method=DeepExMethod, render_speed=DeepExPreset, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "f:\hybrid\64bit\vapoursynth\Lib\site-packages\vsdeoldify\__init__.py", line 1728, in HAVC_deepex clip_colored = vs_colormnet2(clip, clip_ref, clip_sc, image_size=-1, enable_resize=enable_resize, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "f:\hybrid\64bit\vapoursynth\Lib\site-packages\vsdeoldify\vsslib\vsmodels.py", line 70, in vs_colormnet2 return vs_colormnet2_remote(clip, clip_ref, clip_sc, image_size, enable_resize, frame_propagate, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "f:\hybrid\64bit\vapoursynth\Lib\site-packages\vsdeoldify\colormnet2\__init__.py", line 138, in vs_colormnet2_remote clip_colored = _colormnet2_client(colorizer, clip, clip_ref, clip_sc, frame_propagate, ref_weight, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "f:\hybrid\64bit\vapoursynth\Lib\site-packages\vsdeoldify\colormnet2\__init__.py", line 155, in _colormnet2_client reader.load_clip_ref(clip_ref, clip_sc, window_size=max_memory_frames) File "f:\hybrid\64bit\vapoursynth\Lib\site-packages\vsdeoldify\colormnet2\colormnet2_utils.py", line 105, in load_clip_ref self.get_clip_ref_list(sc, start_frame=start_frame, window_size=window_size) File "f:\hybrid\64bit\vapoursynth\Lib\site-packages\vsdeoldify\colormnet2\colormnet2_utils.py", line 92, in get_clip_ref_list HAVC_LogMessage(MessageType.EXCEPTION, File "f:\hybrid\64bit\vapoursynth\Lib\site-packages\vsdeoldify\vsslib\vsutils.py", line 45, in HAVC_LogMessage raise vs.Error(message_text) vapoursynth.Error: CMNET2: number of reference frames must be at least 2, found 2

Cu Selur

Ps.: also uploaded a new Hybrid_havc_test

[Dan64]

Please set ScMinFreq=15, I left some of defaults value of the previous release, CMNET2 need a stable source of reference frames and with the default value of threshold=0.10, it seems that was able to find only 2 reference frames.
I will update the RC.

Dan

[Selur]

But "number of reference frames must be at least 2, found 2" => shouldn't two be enough? Or did you mean that it need '3 or more' `?

Please set ScMinFreq=15, I left some of defaults value of the previous release,

Don't really get that sentence, but using 'ScMinFreq=15':

clip = havc.HAVC_main(clip, ColorFix="none", BlackWhiteTune="none", EnableDeepEx=True, DeepExMethod=0, DeepExPreset="medium", DeepExRefMerge=0, ScFrameDir=None, ScMinFreq=15, ScNormalize=True, DeepExModel=3, DeepExEncMode=0)

doesn't crash.

Let me know when defaults change, so that I can adjust the gui accordingly.

Cu Selur

[Dan64]

Tomorrow I will release a new RC with the updated defaults.

Dan

[Selur]

Okay.

[Dan64]

When is selected the model CMNET2, must be available the same options available for ColorMNet: Mode (remote/local), Frames (number of memory frames).
The Mode="remote all-ref" has been removed in 5.8.0. It was my unsuccessful attempt to implement a sliding memory window.
This approach has been implemented in CMNET2 and thanks to the permanent memory of Xmem++, now is working very well. 

Dan

[Selur]

Uploaded an updated Hybrid_havc_test, let me know how that works.

Cu Selur

[Dan64]

When I select CMNET2 is displayed this panel


 

The option Mode and Frames are missing, in this case should be displayed the same options of ColorMNet as shown below




Dan

[Selur]

Did you use the today updated Hybrid_havc_test?
Because for me it shows:





Cu Selur

[Dan64]

Now it works (probably I unzipped the old version)

I attached the new RC5.

Please check your defaults, the current ones are:

HAVC_main(clip: vs.VideoNode, Preset: str = 'Medium', FrameInterp: int = 0,  ColorModel: str = 'Video+ModelScope', CombMethod: str = 'Simple',  VideoTune: str = 'VeryVivid', ColorFix: str = 'Retinex/Red', ColorTune: str = 'Light', ColorMap: str = 'Red->Brown', ColorTemp: str = "None", BlackWhiteTune: str = 'Light', BlackWhiteMode: int = 0, BlackWhiteBlend: bool = True, EnableDeepEx: bool = True, DeepExMethod: int = 0, DeepExPreset: str = 'Auto', DeepExRefMerge: int = 0, DeepExOnlyRefFrames: bool = False, ScFrameDir: str = None, ScThreshold: float = 0.10, ScThtOffset: int = 1, ScMinFreq: int = 15, ScMinInt: int = 1, ScThtSSIM: float = 0.0, ScNormalize: bool = False, DeepExModel: int = 0, DeepExVivid: bool = False, DeepExEncMode: int = 0, DeepExMaxMemFrames=20, RefRange: tuple[int, int] = (0, 0), enable_fp16: bool = True, debug_level: int = 0)

 The main change is that now the section "Exemplar Models" is enabled by default and configured to use CMNET2.

Dan