BasicVSR++: Improving Video Super-Resolution with Enhanced Propagation and Alignment

Overview

BasicVSR++

BasicVSR++: Improving Video Super-Resolution with Enhanced Propagation and Alignment

Ported from https://github.com/open-mmlab/mmediting

Dependencies

Installing mmcv-full on Windows is a bit complicated as it requires Visual Studio and other tools to compile CUDA ops. So I have uploaded the built file compiled with CUDA 11.1 for Windows users and you can install it by executing the following command.

pip install https://github.com/HolyWu/vs-basicvsrpp/releases/download/v1.0.0/mmcv_full-1.3.12-cp39-cp39-win_amd64.whl

Installation

pip install --upgrade vsbasicvsrpp
python -m vsbasicvsrpp

Usage

from vsbasicvsrpp import BasicVSRPP

ret = BasicVSRPP(clip)

See __init__.py for the description of the parameters.

Comments
  • Question about the tiling,...

    Question about the tiling,...

    I got a Geforce GTX 1070ti with 8 GB of vram. (I know it's not new and really that suited for this, but that's what I got. :)) If I crop my source to a 480x480 chunk and run BasicVSR++ on it ~1.4-3.5GB VRAM are used. Not cropping my source, I thought that even with some padding I thought that using "BasicVSRPP(clip=clip, model=5, tile_x=480, tile_y=480)" would allow me to filter HD an UHD clips which it does not. Question is: Why? Shouldn't this work with a 480x480 tiling?

    What are the min tile with and height sizes (I assumed it would be 64*4=256 + padding, but when using 320x320 I get: Python exception: Analyse: failed to retrieve first frame from super clip. Error message: The height and width of low-res inputs must be at least 64, but got 84 and 44. Using 392x392 I get The height and width of low-res inputs must be at least 64, but got 102 and 26. ->Just from testing I don't get what the tiling does at all. :)

    opened by Selur 5
  • Question regarding some Arguments

    Question regarding some Arguments

    Hi, My appologies, I'm not sure if I forgot to submit my last typed issue or if it was removed. If it was removed, feel free to close this without a comment. I typed it this morning, and honestly, can't remember if I submitted or closed the browser window by accident.

    I have a few questions for some of the arguments one can set, as those are a bit different from the mmediting interface, if I have seen that correctly.

    Interval: This specifies the number of Images per Batch, is that correct? In addition of reducing the VRam footprint, I assume it influences what the network can see during the upscale process? The smaller the batch, the smaller the window it can include in the temporal calculations?

    Tiling: This splits the image into tiles for processing instead of the whole image, is that correct? What is the purpose or best case scenario someone would use this for?

    FP16: I run the network with FP16 at the moment, as FP32 usually blows up my 11GB of VRam if I don't reduce the Interval. As I had not enough time to finish all my Tests yesterday, have you noticed any degradation in quality in Images when FP16 is used? If I want to run it with FP32, I need to reduce the Interval but if my assumption is correct, this narrows down the time-window the network can calculate across. So I'm in between Precision vs Intervall-Size for VRam usage. Do you have any experience in this and what a good tradeoff might look like.

    opened by Memnarch 4
  • mmcv-full on Windows and Python 3.10

    mmcv-full on Windows and Python 3.10

    Since Vapoursynth RC58 either need Python 3.8 (Win7 compatible) or Python 3.10 (which I'm using), the current mmcv_full-1.3.16-cp39-cp39-win_amd64.whl is not a supported wheel since it's meant for Python 3.9. Would be nice if you could create a mmcv-full for Python 3.10 in Windows. Thanks!

    opened by Selur 2
  • Sizes of tensors must match except in dimension 3. Got 213 and 214 (The offending index is 0)

    Sizes of tensors must match except in dimension 3. Got 213 and 214 (The offending index is 0)

    python 3.9.7 torch 1.9.1 mmcv 1.3.13/1.4.4 vs r57 vs fatpack

    import vapoursynth as vs core = vs.core

    from vsbasicvsrpp import BasicVSRPP

    video = core.ffms2.Source(source=r'D:\winpython\VapourSynth64Portable\in.mp4')

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

    video = BasicVSRPP(clip=video, interval=10, model=5, fp16=True, tile_pad=0)

    video = core.resize.Bicubic(video, format=vs.YUV420P8, matrix_s="709")

    video.set_output()

    opened by oblessnoob 2
  • meshgrid() got an unexpected keyword argument 'indexing'

    meshgrid() got an unexpected keyword argument 'indexing'

    Using the v1.4.0 and: clip = BasicVSRPP(clip=clip, model=3, tile_x=352, tile_y=480, fp16=True) full script:

    # Imports
    import vapoursynth as vs
    # getting Vapoursynth core
    core = vs.core
    # Loading Plugins
    core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/DeinterlaceFilter/TIVTC/libtivtc.dll")
    core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/SourceFilter/d2vSource/d2vsource.dll")
    # source: 'C:\Users\Selur\Desktop\VTS_02_1-Sample-Beginning.demuxed.m2v'
    # current color space: YUV420P8, bit depth: 8, resolution: 720x480, fps: 29.97, color matrix: 470bg, yuv luminance scale: limited, scanorder: telecine
    # Loading C:\Users\Selur\Desktop\VTS_02_1-Sample-Beginning.demuxed.m2v using D2VSource
    clip = core.d2v.Source(input="E:/Temp/m2v_154f3f0f52f994b09117b9c8650e17d2_853323747.d2v")
    # making sure input color matrix is set as 470bg
    clip = core.resize.Bicubic(clip, matrix_in_s="470bg",range_s="limited")
    # making sure frame rate is set to 29.97
    clip = core.std.AssumeFPS(clip=clip, fpsnum=30000, fpsden=1001)
    # Setting color range to TV (limited) range.
    clip = core.std.SetFrameProp(clip=clip, prop="_ColorRange", intval=1)
    # Deinterlacing using TIVTC
    clip = core.tivtc.TFM(clip=clip)
    clip = core.tivtc.TDecimate(clip=clip)# new fps: 23.976
    # make sure content is preceived as frame based
    clip = core.std.SetFieldBased(clip, 0)
    # DEBUG: vsTIVTC changed scanorder to: progressive
    # cropping the video to 704x480
    clip = core.std.CropRel(clip=clip, left=6, right=10, top=0, bottom=0)
    # adjusting color space from YUV420P8 to RGBS for vsBasicVSRPPFilter
    clip = core.resize.Bicubic(clip=clip, format=vs.RGBS, matrix_in_s="470bg", range_s="limited")
    # Quality enhancement using BasicVSR++
    from vsbasicvsrpp import BasicVSRPP
    clip = BasicVSRPP(clip=clip, model=3, tile_x=352, tile_y=480, fp16=True)
    # adjusting output color from: RGBS to YUV420P8 for x264Model
    clip = core.resize.Bicubic(clip=clip, format=vs.YUV420P8, matrix_s="470bg", range_s="limited")
    # set output frame rate to 23.976fps
    clip = core.std.AssumeFPS(clip=clip, fpsnum=24000, fpsden=1001)
    # Output
    clip.set_output()
    

    I get:

    Error on frame 0 request:
    meshgrid() got an unexpected keyword argument 'indexing'
    
    opened by Selur 2
  • ImportError: DLL load failed while importing _ext: The specified procedure could not be found.

    ImportError: DLL load failed while importing _ext: The specified procedure could not be found.

    Using Python 3.9.7, I'm having the following issue when I try to finish installing vsbasicvsrpp with python -m vsbasicvsrpp.

    I had installed it just fine weeks ago, but I attempted to update PyTorch last night with: pip3 install torch==1.10.0+cu113 torchvision==0.11.1+cu113 -f https://download.pytorch.org/whl/cu113/torch_stable.html and that seemed to break everything, so I managed to fix it, but vsbasicvsrpp remains unfixed due to the following.

    (c) Microsoft Corporation. All rights reserved.
    
    C:\WINDOWS\system32>pip install --upgrade vsbasicvsrpp
    WARNING: Ignoring invalid distribution mvtools-float- (c:\users\mainuser\appdata\roaming\python\python39\site-packages)
    WARNING: Ignoring invalid distribution mvtools-float- (c:\users\mainuser\appdata\roaming\python\python39\site-packages)
    Collecting vsbasicvsrpp
      Using cached vsbasicvsrpp-1.3.0-py3-none-any.whl (21 kB)
    Requirement already satisfied: torchvision in c:\users\mainuser\appdata\local\programs\python\python39\lib\site-packages (from vsbasicvsrpp) (0.11.1+cu113)
    Requirement already satisfied: mmcv-full>=1.3.13 in c:\users\mainuser\appdata\local\programs\python\python39\lib\site-packages (from vsbasicvsrpp) (1.3.14)
    Requirement already satisfied: torch>=1.9.0 in c:\users\mainuser\appdata\local\programs\python\python39\lib\site-packages (from vsbasicvsrpp) (1.10.0+cu113)
    Requirement already satisfied: numpy in c:\users\mainuser\appdata\local\programs\python\python39\lib\site-packages (from vsbasicvsrpp) (1.21.3)
    Requirement already satisfied: Pillow in c:\users\mainuser\appdata\local\programs\python\python39\lib\site-packages (from mmcv-full>=1.3.13->vsbasicvsrpp) (8.3.1)
    Requirement already satisfied: packaging in c:\users\mainuser\appdata\local\programs\python\python39\lib\site-packages (from mmcv-full>=1.3.13->vsbasicvsrpp) (21.0)
    Requirement already satisfied: addict in c:\users\mainuser\appdata\local\programs\python\python39\lib\site-packages (from mmcv-full>=1.3.13->vsbasicvsrpp) (2.4.0)
    Requirement already satisfied: yapf in c:\users\mainuser\appdata\local\programs\python\python39\lib\site-packages (from mmcv-full>=1.3.13->vsbasicvsrpp) (0.31.0)
    Requirement already satisfied: regex in c:\users\mainuser\appdata\local\programs\python\python39\lib\site-packages (from mmcv-full>=1.3.13->vsbasicvsrpp) (2021.8.28)
    Requirement already satisfied: pyyaml in c:\users\mainuser\appdata\local\programs\python\python39\lib\site-packages (from mmcv-full>=1.3.13->vsbasicvsrpp) (5.4.1)
    Requirement already satisfied: typing-extensions in c:\users\mainuser\appdata\local\programs\python\python39\lib\site-packages (from torch>=1.9.0->vsbasicvsrpp) (3.10.0.0)
    Requirement already satisfied: pyparsing>=2.0.2 in c:\users\mainuser\appdata\local\programs\python\python39\lib\site-packages (from packaging->mmcv-full>=1.3.13->vsbasicvsrpp) (2.4.7)
    WARNING: Ignoring invalid distribution mvtools-float- (c:\users\mainuser\appdata\roaming\python\python39\site-packages)
    Installing collected packages: vsbasicvsrpp
    Successfully installed vsbasicvsrpp-1.3.0
    WARNING: Ignoring invalid distribution mvtools-float- (c:\users\mainuser\appdata\roaming\python\python39\site-packages)
    WARNING: Ignoring invalid distribution mvtools-float- (c:\users\mainuser\appdata\roaming\python\python39\site-packages)
    
    C:\WINDOWS\system32>python -m vsbasicvsrpp
    Traceback (most recent call last):
      File "C:\Users\MainUser\AppData\Local\Programs\Python\Python39\lib\runpy.py", line 188, in _run_module_as_main
        mod_name, mod_spec, code = _get_module_details(mod_name, _Error)
      File "C:\Users\MainUser\AppData\Local\Programs\Python\Python39\lib\runpy.py", line 147, in _get_module_details
        return _get_module_details(pkg_main_name, error)
      File "C:\Users\MainUser\AppData\Local\Programs\Python\Python39\lib\runpy.py", line 111, in _get_module_details
        __import__(pkg_name)
      File "C:\Users\MainUser\AppData\Local\Programs\Python\Python39\lib\site-packages\vsbasicvsrpp\__init__.py", line 10, in <module>
        from .basicvsr_pp import BasicVSRPlusPlus
      File "C:\Users\MainUser\AppData\Local\Programs\Python\Python39\lib\site-packages\vsbasicvsrpp\basicvsr_pp.py", line 8, in <module>
        from mmcv.ops import ModulatedDeformConv2d, modulated_deform_conv2d
      File "C:\Users\MainUser\AppData\Local\Programs\Python\Python39\lib\site-packages\mmcv\ops\__init__.py", line 2, in <module>
        from .ball_query import ball_query
      File "C:\Users\MainUser\AppData\Local\Programs\Python\Python39\lib\site-packages\mmcv\ops\ball_query.py", line 7, in <module>
        ext_module = ext_loader.load_ext('_ext', ['ball_query_forward'])
      File "C:\Users\MainUser\AppData\Local\Programs\Python\Python39\lib\site-packages\mmcv\utils\ext_loader.py", line 13, in load_ext
        ext = importlib.import_module('mmcv.' + name)
      File "C:\Users\MainUser\AppData\Local\Programs\Python\Python39\lib\importlib\__init__.py", line 127, in import_module
        return _bootstrap._gcd_import(name[level:], package, level)
    ImportError: DLL load failed while importing _ext: The specified procedure could not be found.
    opened by AIisCool 2
  • Adjust strength?

    Adjust strength?

    First off I wish to say thank you very much for all the work you do HolyWu!

    I love how much noise vs-basicvsrpp removes from the video, but sometimes it's a little too much (ie. removes something that isn't noise at all). Is there any way to adjust the strength of it similar to DPIR?

    Thank you.

    opened by AIisCool 2
  • Model 5 seems to downscale by 4

    Model 5 seems to downscale by 4

    Hi, While I got good results, they did not seem to be quite as sharp as I expected them. When I tried to process a 136x264 Video, I suddenly got the error:

    Traceback (most recent call last):
      File "E:\Git\RivenTools\Reproduce.py", line 26, in <module>
        video.output(f, y4m=True)
      File "src\cython\vapoursynth.pyx", line 1790, in vapoursynth.VideoNode.output
      File "src\cython\vapoursynth.pyx", line 1655, in frames
      File "D:\Program Files\Python39\lib\concurrent\futures\_base.py", line 445, in result
        return self.__get_result()
      File "D:\Program Files\Python39\lib\concurrent\futures\_base.py", line 390, in __get_result
        raise self._exception
    vapoursynth.Error: The height and width of low-res inputs must be at least 64, but got 66 and 34.
    

    But my Video is larger. Looking into it, executing Model 5 downscales the video. Or at least its internal data seems to be downscaled. While the returned clip reports the correct size, the frames seem to be lower. A quarter to be precise.

    This seems to be the code causing it: https://github.com/HolyWu/vs-basicvsrpp/blob/da066461f66c6e7deedf354630899b815393836b/vsbasicvsrpp/basicvsr_pp.py#L291

    And this is the script to reproduce it (It's a trimmed down version, that's why model 5 and 1 are executed directly after one another. My pipeline has some steps inbetween): Reproduce.zip

    If you need that specific video from my script, I can share that with you but would prefer to do that outside of this report, as it's a game asset.

    opened by Memnarch 2
  • got a warning,...

    got a warning,...

    Running vs-basicvsrpp I get:

    I:\Hybrid\64bit\Vapoursynth\Lib\site-packages\torch\nn\functional.py:3657: UserWarning: The default behavior for interpolate/upsample with float scale_factor changed in 1.6.0 to align with other frameworks/libraries, and now uses scale_factor directly, instead of relying on the computed output size. If you wish to restore the old behavior, please set recompute_scale_factor=True. See the documentation of nn.Upsample for details.
      warnings.warn(
    

    should I do something or should I ignore this?

    opened by Selur 2
  • calling

    calling "python -m vsbasicvsrpp" fails

    Using a portable Vapoursynth R58 and calling python -m pip install --upgrade vsbasicvsrpp gives me

    Requirement already satisfied: vsbasicvsrpp in i:\hybrid\64bit\vapoursynth\lib\site-packages (1.4.1)
    Requirement already satisfied: torchvision in i:\hybrid\64bit\vapoursynth\lib\site-packages (from vsbasicvsrpp) (0.12.0)
    Requirement already satisfied: mmcv-full>=1.3.13 in i:\hybrid\64bit\vapoursynth\lib\site-packages (from vsbasicvsrpp) (1.5.0)
    Requirement already satisfied: numpy in i:\hybrid\64bit\vapoursynth\lib\site-packages (from vsbasicvsrpp) (1.22.3)
    Requirement already satisfied: torch>=1.9.0 in i:\hybrid\64bit\vapoursynth\lib\site-packages (from vsbasicvsrpp) (1.11.0)
    Requirement already satisfied: regex in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full>=1.3.13->vsbasicvsrpp) (2022.3.15)
    Requirement already satisfied: pyyaml in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full>=1.3.13->vsbasicvsrpp) (6.0)
    Requirement already satisfied: packaging in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full>=1.3.13->vsbasicvsrpp) (21.3)
    Requirement already satisfied: Pillow in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full>=1.3.13->vsbasicvsrpp) (9.1.0)
    Requirement already satisfied: addict in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full>=1.3.13->vsbasicvsrpp) (2.4.0)
    Requirement already satisfied: yapf in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full>=1.3.13->vsbasicvsrpp) (0.32.0)
    Requirement already satisfied: typing-extensions in i:\hybrid\64bit\vapoursynth\lib\site-packages (from torch>=1.9.0->vsbasicvsrpp) (4.2.0)
    Requirement already satisfied: requests in i:\hybrid\64bit\vapoursynth\lib\site-packages (from torchvision->vsbasicvsrpp) (2.27.1)
    Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in i:\hybrid\64bit\vapoursynth\lib\site-packages (from packaging->mmcv-full>=1.3.13->vsbasicvsrpp) (3.0.8)
    Requirement already satisfied: idna<4,>=2.5 in i:\hybrid\64bit\vapoursynth\lib\site-packages (from requests->torchvision->vsbasicvsrpp) (3.3)
    Requirement already satisfied: certifi>=2017.4.17 in i:\hybrid\64bit\vapoursynth\lib\site-packages (from requests->torchvision->vsbasicvsrpp) (2021.10.8)
    Requirement already satisfied: urllib3<1.27,>=1.21.1 in i:\hybrid\64bit\vapoursynth\lib\site-packages (from requests->torchvision->vsbasicvsrpp) (1.26.9)
    Requirement already satisfied: charset-normalizer~=2.0.0 in i:\hybrid\64bit\vapoursynth\lib\site-packages (from requests->torchvision->vsbasicvsrpp) (2.0.12)
    

    which looks fine to me. Problem is, calling python -m vsbasicvsrpp gives me:

    Traceback (most recent call last):
      File "runpy.py", line 187, in _run_module_as_main
      File "runpy.py", line 146, in _get_module_details
      File "runpy.py", line 110, in _get_module_details
      File "i:\Hybrid\64bit\Vapoursynth\Lib\site-packages\vsbasicvsrpp\__init__.py", line 10, in <module>
        from .basicvsr_pp import BasicVSRPlusPlus
      File "i:\Hybrid\64bit\Vapoursynth\Lib\site-packages\vsbasicvsrpp\basicvsr_pp.py", line 8, in <module>
        from mmcv.ops import ModulatedDeformConv2d, modulated_deform_conv2d
      File "i:\Hybrid\64bit\Vapoursynth\Lib\site-packages\mmcv\ops\__init__.py", line 2, in <module>
        from .active_rotated_filter import active_rotated_filter
      File "i:\Hybrid\64bit\Vapoursynth\Lib\site-packages\mmcv\ops\active_rotated_filter.py", line 8, in <module>
        ext_module = ext_loader.load_ext(
      File "i:\Hybrid\64bit\Vapoursynth\Lib\site-packages\mmcv\utils\ext_loader.py", line 13, in load_ext
        ext = importlib.import_module('mmcv.' + name)
      File "importlib\__init__.py", line 126, in import_module
    

    any idea what I'm doing wrong / where the problem is?

    Calling python -m pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cu113/torch1.11/index.html gives me:

    Looking in links: https://download.openmmlab.com/mmcv/dist/cu113/torch1.11/index.html
    Requirement already satisfied: mmcv-full in i:\hybrid\64bit\vapoursynth\lib\site-packages (1.5.0)
    Requirement already satisfied: pyyaml in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full) (6.0)
    Requirement already satisfied: numpy in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full) (1.22.3)
    Requirement already satisfied: addict in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full) (2.4.0)
    Requirement already satisfied: Pillow in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full) (9.1.0)
    Requirement already satisfied: yapf in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full) (0.32.0)
    Requirement already satisfied: regex in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full) (2022.3.15)
    Requirement already satisfied: packaging in i:\hybrid\64bit\vapoursynth\lib\site-packages (from mmcv-full) (21.3)
    Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in i:\hybrid\64bit\vapoursynth\lib\site-packages (from packaging->mmcv-full) (3.0.8)
    

    which seems fine too, to me.

    Cu Selur

    opened by Selur 1
  • NTIRE 2021 Video Super-Resolution

    NTIRE 2021 Video Super-Resolution

    opened by AIisCool 1
  • Multi-gpu support?

    Multi-gpu support?

    I tried to utilize 4 gpus for inference with the following code but it didn't work. Only one of the gpu was doing the job at a time and the others were idling. Are there any suggested ways for multi-gpu inference?

    import vapoursynth as vs
    import os
    from vsbasicvsrpp import BasicVSRPP
    core = vs.core
    
    folder = r'C:\Users\test\Desktop\vs-58'
    file = r'test.m4v'
    
    src = os.path.join(folder, file)
    
    src = core.ffms2.Source(src)
    
    src = core.fmtc.resample (clip=src, css="444")
    src = core.fmtc.matrix (clip=src, mat="709", col_fam=vs.RGB)
    src = core.fmtc.bitdepth (clip=src, bits=32)
    
    interval = 180
    n = 4
    
    add = (interval*n) - len(src) % (interval*n)
    if add>0:
    	src = src + core.std.BlankClip(src, length=add)
    
    c1 = core.std.SelectEvery(clip=src, cycle=interval*n,  offsets=[i for i in range(interval*0, interval)])
    c1 = BasicVSRPP(c1, model=5, interval=interval, device_index=0, fp16=True)
    
    c2 = core.std.SelectEvery(clip=src, cycle=interval*n,  offsets=[i for i in range(interval*1, interval*2)])
    c2 = BasicVSRPP(c2, model=5, interval=interval, device_index=1, fp16=True)
    
    c3 = core.std.SelectEvery(clip=src, cycle=interval*n,  offsets=[i for i in range(interval*2, interval*3)])
    c3 = BasicVSRPP(c3, model=5, interval=interval, device_index=2, fp16=True)
    
    c4 = core.std.SelectEvery(clip=src, cycle=interval*n,  offsets=[i for i in range(interval*3, interval*4)])
    c4 = BasicVSRPP(c4, model=5, interval=interval, device_index=3, fp16=True)
    
    
    c = core.std.Interleave(clips=[c1, c2, c3, c4])
    
    a = [i for i in range(interval*n) if i % n ==0] + [i for i in range(interval*n) if i % n ==1] + [i for i in range(interval*n) if i % n ==2] + [i for i in range(interval*n) if i % n ==3]
    
    c = core.std.SelectEvery(clip=c, cycle=interval*n, offsets=a)
    
    c = core.fmtc.matrix (clip=c, mat="709", col_fam=vs.YUV)
    c = core.fmtc.resample (clip=c, css="420")
    c = core.fmtc.bitdepth(clip =c, bits=16)   
    
    if add>0:
    	c = c[:-add]
    
    c.set_output()
    
    opened by Bouby308 0
  • deepfillv2 support

    deepfillv2 support

    thanks for your effort:smile: At present, there is't a good inpaint method in vs. i notice mmediting also support deepfillv2, how about port it to vs (it performs good for image, maybe temporal flicker for video)

    opened by soldivelot 0
Releases(v1.4.1)
Owner
Holy Wu
Holy Wu
Progressive Growing of GANs for Improved Quality, Stability, and Variation

Progressive Growing of GANs for Improved Quality, Stability, and Variation — Official TensorFlow implementation of the ICLR 2018 paper Tero Karras (NV

Tero Karras 5.9k Jan 05, 2023
Official repository for MixFaceNets: Extremely Efficient Face Recognition Networks

MixFaceNets This is the official repository of the paper: MixFaceNets: Extremely Efficient Face Recognition Networks. (Accepted in IJCB2021) https://i

Fadi Boutros 51 Dec 13, 2022
Abstractive opinion summarization system (SelSum) and the largest dataset of Amazon product summaries (AmaSum). EMNLP 2021 conference paper.

Learning Opinion Summarizers by Selecting Informative Reviews This repository contains the codebase and the dataset for the corresponding EMNLP 2021

Arthur Bražinskas 39 Jan 01, 2023
領域を指定し、キーを入力することで画像を保存するツールです。クラス分類用のデータセット作成を想定しています。

image-capture-class-annotation 領域を指定し、キーを入力することで画像を保存するツールです。 クラス分類用のデータセット作成を想定しています。 Requirement OpenCV 3.4.2 or later Usage 実行方法は以下です。 起動後はマウスクリック4

KazuhitoTakahashi 5 May 28, 2021
Pytorch implementation of the DeepDream computer vision algorithm

deep-dream-in-pytorch Pytorch (https://github.com/pytorch/pytorch) implementation of the deep dream (https://en.wikipedia.org/wiki/DeepDream) computer

102 Dec 05, 2022
Contrastive Learning for Many-to-many Multilingual Neural Machine Translation(mCOLT/mRASP2), ACL2021

Contrastive Learning for Many-to-many Multilingual Neural Machine Translation(mCOLT/mRASP2), ACL2021 The code for training mCOLT/mRASP2, a multilingua

104 Jan 01, 2023
Tensorflow 2.x implementation of Panoramic BlitzNet for object detection and semantic segmentation on indoor panoramic images.

Deep neural network for object detection and semantic segmentation on indoor panoramic images. The implementation is based on the papers:

Alejandro de Nova Guerrero 9 Nov 24, 2022
PASTRIE: A Corpus of Prepositions Annotated with Supersense Tags in Reddit International English

PASTRIE Official release of the corpus described in the paper: Michael Kranzlein, Emma Manning, Siyao Peng, Shira Wein, Aryaman Arora, and Nathan Schn

NERT @ Georgetown 4 Dec 02, 2021
App for identification of various objects. Based on YOLO v4 tiny architecture

Object_detection Repository containing trained model yolo v4 tiny, which is capable of identification 80 different classes Default feed is set to be a

Mateusz Kurdziel 0 Jun 22, 2022
PyTorch Implementation of Temporal Output Discrepancy for Active Learning, ICCV 2021

Temporal Output Discrepancy for Active Learning PyTorch implementation of Semi-Supervised Active Learning with Temporal Output Discrepancy, ICCV 2021.

Siyu Huang 33 Dec 06, 2022
Fine-grained Post-training for Improving Retrieval-based Dialogue Systems - NAACL 2021

Fine-grained Post-training for Multi-turn Response Selection Implements the model described in the following paper Fine-grained Post-training for Impr

Janghoon Han 83 Dec 20, 2022
MODNet: Trimap-Free Portrait Matting in Real Time

MODNet is a model for real-time portrait matting with only RGB image input.

Zhanghan Ke 2.8k Dec 30, 2022
The code for our NeurIPS 2021 paper "Kernelized Heterogeneous Risk Minimization".

Kernelized-HRM Jiashuo Liu, Zheyuan Hu The code for our NeurIPS 2021 paper "Kernelized Heterogeneous Risk Minimization"[1]. This repo contains the cod

Liu Jiashuo 8 Nov 20, 2022
Wind Speed Prediction using LSTMs in PyTorch

Implementation of Deep-Forecast using PyTorch Deep Forecast: Deep Learning-based Spatio-Temporal Forecasting Adapted from original implementation Setu

Onur Kaplan 151 Dec 14, 2022
A package for "Procedural Content Generation via Reinforcement Learning" OpenAI Gym interface.

Readme: Illuminating Diverse Neural Cellular Automata for Level Generation This is the codebase used to generate the results presented in the paper av

Sam Earle 27 Jan 05, 2023
SwinIR: Image Restoration Using Swin Transformer

SwinIR: Image Restoration Using Swin Transformer This repository is the official PyTorch implementation of SwinIR: Image Restoration Using Shifted Win

Jingyun Liang 2.4k Jan 05, 2023
[NeurIPS 2021] Better Safe Than Sorry: Preventing Delusive Adversaries with Adversarial Training

Better Safe Than Sorry: Preventing Delusive Adversaries with Adversarial Training Code for NeurIPS 2021 paper "Better Safe Than Sorry: Preventing Delu

Lue Tao 29 Sep 20, 2022
Learning embeddings for classification, retrieval and ranking.

StarSpace StarSpace is a general-purpose neural model for efficient learning of entity embeddings for solving a wide variety of problems: Learning wor

Facebook Research 3.8k Dec 22, 2022
This repository is all about spending some time the with the original problem posed by Minsky and Papert

This repository is all about spending some time the with the original problem posed by Minsky and Papert. Working through this problem is a great way to begin learning computer vision.

Jaissruti Nanthakumar 1 Jan 23, 2022
Context Axial Reverse Attention Network for Small Medical Objects Segmentation

CaraNet: Context Axial Reverse Attention Network for Small Medical Objects Segmentation This repository contains the implementation of a novel attenti

401 Dec 23, 2022