A High-Quality Real Time Upscaler for Anime Video

Overview

Anime4K

Anime4K is a set of open-source, high-quality real-time anime upscaling/denoising algorithms that can be implemented in any programming language.

The simplicity and speed of Anime4K allows the user to watch upscaled anime in real time, as we believe in preserving original content and promoting freedom of choice for all anime fans. Re-encoding anime into 4K should be avoided as it is non-reversible, potentially damages original content by introducing artifacts, takes up to O(n2) more disk space and more importantly, does so without any meaningful decrease in entropy (lost information is lost).

Disclaimer: All art assets used are for demonstration and educational purposes. All rights are reserved to their original owners. If you (as a person or a company) own the art and do not wish it to be associated with this project, please contact us at [email protected] and we will gladly take it down.

Foreword

Anime4K is optimized for native 1080p anime encoded with h.264, h.265 or VC-1.

Even if it might work, it is not optimized for downscaled 720p, 480p or standard definition anime (eg. DVDs). Older anime (especially pre-digital era production) have artifacts that are very difficult to remove, such as bad deinterlacing, camera blur during production, severe ringing, film grain, older MPEG compression artifacts, etc.

This is also not replacement for SRGANs, as they perform much better on low-resolution images or images with lots of degradation (albeit not in real time).

What Anime4K does provide is a way to upscale, in real time, 1080p anime for 4K screens while providing a similar effect to SRGANs and being much better than waifu2x (See comparisons).

Currently, research is being done on better real-time upscaling for lower resolution or older content.

v4.1 Low resolution experiment

Results from the experimental SRGAN shaders for 360p -> 4K: (zoom in to view details)

The images are sorted by algorithm speed, bicubic being the fastest. FSRCNNX and Anime4K are real-time while waifu2x and Real-ESRGAN are not. Comparison Comparison

v4

Installation Instructions for GLSL/MPV

We introduce a line reconstruction algorithm that aims to tackle the distribution shift problem seen in 1080p anime. In the wild anime exhibit a surprising amount of variance caused by low quality compositing due to budget and time constraints that traditional super-resolution algorithms cannot handle. GANs can implicitly encode this distribution shift but are slow to use and hard to train. Our algorithm explicitly corrects this distribution shift and allows traditional "MSE" SR algorithms to work with a wide variety of anime.

Source: https://fancaps.net/anime/picture.php?/14728493 | Mode: B
Comparison

Source: https://fancaps.net/anime/picture.php?/13365760 | Mode: A
Comparison

Performance numbers are obtained using a Vega64 GPU and are tested using UL shader variants. The fast version is for M variants.
Note that CUDA accelerated SRGANs/Waifu2x using tensor cores can be much faster and close to realtime (~80ms), but their large size severely hampers non-CUDA implementations.

v3

The monolithic Anime4K shader is broken into modular components, allowing customization for specific types of anime and/or personal taste. What's new:

  • A complete overhaul of the algorithm(s) for speed, quality and efficiency.
  • Real-time, high quality line art CNN upscalers. (6 variants)
  • Line art deblurring shaders. ("blind deconvolution" and DTD shader)
  • Denoising algorithms. (Bilateral Mode and CNN variants)
  • Blind resampling artifact reduction algorithms. (For badly resampled anime.)
  • Experimental line darkening and line thinning algorithm. (For perceptual quality. We perceive thinner/darker lines as perceptually higher quality, even if it might not be the case.)

More information about each shader (OUTDATED).

Visits

Counting since 2021-09-19T16:02:06Z (ISO 8601)

Projects that use Anime4K

Note that the following might be using an outdated version of Anime4K. There have been significant quality improvements since v3.

Acknowledgements

OpenCV TensorFlow Keras Torch mpv MPC
OpenCV TensorFlow Keras Torch mpv MPC

Many thanks to the OpenCV, TensorFlow, Keras and Torch groups and contributors. This project would not have been possible without the existence of high quality, open source machine learning libraries.

I would also want to specially thank the creators of VDSR and FSRCNN, in addition to the open source projects waifu2x and FSRCNNX for sparking my interest in creating this project. I am also extending my gratitude to the contributors of mpv and MPC-HC/BE for their efforts on creating excellent media players with endless customization options.
Furthermore, I want to thank the people who contributed to this project in any form, be it by reporting bugs, submitting suggestions, helping others' issues or submitting code. I will forever hold you in high regard.

I also wish to express my sincere gratitude to the people of Université de Montréal, DIRO, LIGUM and MILA for providing so many opportunities to students (including me), providing the necessary infrastructure and fostering an excellent learning environment.

I would also like to thank the greater open source community, in which the assortment of concrete examples and code were of great help.

Finally, but not least, infinite thanks to my family, friends and professors for providing financial, technical, social support and expertise for my ongoing learning journey during these hard times. Your help has been beyond description, really.

This list is not final, as the project is far from done. Any future acknowledgements will be promptly added.

Comments
  • Solved: Strange performance hit with 480p videos

    Solved: Strange performance hit with 480p videos

    The Nvidia 1050 Ti video card

    I got 60 fps while playing 720 and 1080 videos, only with 480 videos
    the fps drop to around 42 at all time

    Change MPV player scale algorithm to nearest to see if this is the MPV issue but not helped.

    Solved(2020/04/27):

    The reason is turned on video fps interpolation by a misunderstanding.
    True problem is sometimes the video behave discontinuous or drop frames.
    Because video is 23.976 fps but monitor running on 60 fps.

    Interpolation not just convert the frame interval but try to smooth between them.
    This added a lot of resource consume. If just want to avoid frame interval problem
    and not care about smooth, remove all interpolation options and add
    video-sync=display-resample. It's almost zero performance cost.

    opened by mhtvsSFrpHdE 11
  • Will there be an update for the hlsl version?

    Will there be an update for the hlsl version?

    Hi, I have been using your algorithm on top of NGU for the past 3 months or so. I have noticed that the hlsl version has not been updated in quite a while. I tried to convert the glsl shader to hlsl using glslang and SPIRV-Cross but I failed. I would like to know if you are going to update the hlsl version in the future or if there is an easier way to convert the glsl shader.

    opened by Artins90 10
  • 2 questions

    2 questions

    1
    https://github.com/bloc97/Anime4K/blob/master/Java_Instructions.md how do you get the default upsampling scale first, by bicubic? or bilinear?of other methods? And then use the enhancement Anime4K method.

    2
    https://github.com/bloc97/Anime4K/blob/master/results/Graph.png?raw=true how do you get the perceptual quality?

    opened by splinter21 9
  • Anime4K GAN shaders cause video to freeze

    Anime4K GAN shaders cause video to freeze

    Describe the bug I added the following line to my mpv input.conf

    CTRL+8 no-osd change-list glsl-shaders set "~~/shaders/Anime4K_Restore_GAN_UUL.glsl:~~/shaders/Anime4K_Upscale_GAN_x4_UUL.glsl:~~/shaders/Anime4K_Restore_CNN_Soft_M.glsl:~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl"; show-text "Anime4K: Mode GAN (HQ)" Upon pressing CTRL+8, the video freezes while audio continues as normal. I am not even able to control mpv as normal and instead have to kill the process on the terminal.

    To Reproduce Steps to reproduce the behavior:

    1. Add the above line to your mpv config.
    2. Play a low res anime show
    3. Activate the shader
    4. See error

    Expected behavior I expected the video to be upscaled to some degree.

    Screenshots Unable to take screenshots as the display just completely freezes.

    Desktop (please complete the following information):

    • OS: openSUSE Tumbleweed
    • Version: Latest snapshot
    • GPU/Version: Asus RX 6900XT with the current GPU drivers
    • Media Player and Version: mpv v0.34

    Additional context This was done using the latest pull of the master branch of Anime4k.

    An semi-blindly edited config based on mode A with the CNN stages replaced with GAN seems to work, although the results are not as nice as I'd want.

    CTRL+7 no-osd change-list glsl-shaders set "~~/shaders/Anime4K_Clamp_Highlights.glsl:~~/shaders/Anime4K_Restore_GAN_UL.glsl:~~/shaders/Anime4K_Upscale_GAN_x2_M.glsl:~~/shaders/Anime4K_AutoDownscalePre_x2.glsl:~~/shaders/Anime4K_AutoDownscalePre_x4.glsl:~~/shaders/Anime4K_Upscale_GAN_x2_M.glsl"; show-text "Anime4K: Mode GAN"

    opened by duelistgamer 8
  • Anime4K loads in, but shaders dont seem to apply[Linux 64-bit, Nvidia]

    Anime4K loads in, but shaders dont seem to apply[Linux 64-bit, Nvidia]

    Describe the bug Upon following the installation guide mentioned in the wiki, I expected the video to improve it's quality, however, SHIFT+I doesn't show anything.

    I have to also mention the fact that I have also replaced the ";" with ":"

    Although, mpv does seem to recognize that I do have Anime4K, as the keybinds clearly work.

    To Reproduce Follow the instructions mentioned in wiki

    Expected behavior Expected the shaders to apply

    Screenshots Herein lie code blocks and screenshots

    mpv.conf

    sub-auto=fuzzy
    profile=gpu-hq
    glsl-shaders="~~/shaders/Anime4K_Denoise_Bilateral_Mode.glsl:~~/shaders/Anime4K_Deblur_DoG.glsl:~~/shaders/Anime4K_DarkLines_HQ.glsl:~~/shaders/Anime4K_ThinLines_HQ.glsl:~~/shaders/Anime4K_Upscale_CNN_M_x2_Deblur.glsl"
    

    input.conf (This is extracted directly from the one in the wiki, with adjustments made for UNIX-based systems)

    CTRL+1 no-osd change-list glsl-shaders set "~~/shaders/Anime4K_Clamp_Highlights.glsl:~~/shaders/Anime4K_Restore_CNN_M.glsl:~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl:~~/shaders/Anime4K_AutoDownscalePre_x2.glsl:~~/shaders/Anime4K_AutoDownscalePre_x4.glsl:~~/shaders/Anime4K_Upscale_CNN_x2_S.glsl"; show-text "Anime4K: Mode A (Fast)"
    CTRL+2 no-osd change-list glsl-shaders set "~~/shaders/Anime4K_Clamp_Highlights.glsl:~~/shaders/Anime4K_Restore_CNN_Soft_M.glsl:~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl:~~/shaders/Anime4K_AutoDownscalePre_x2.glsl:~~/shaders/Anime4K_AutoDownscalePre_x4.glsl:~~/shaders/Anime4K_Upscale_CNN_x2_S.glsl"; show-text "Anime4K: Mode B (Fast)"
    CTRL+3 no-osd change-list glsl-shaders set "~~/shaders/Anime4K_Clamp_Highlights.glsl:~~/shaders/Anime4K_Upscale_Denoise_CNN_x2_M.glsl:~~/shaders/Anime4K_AutoDownscalePre_x2.glsl:~~/shaders/Anime4K_AutoDownscalePre_x4.glsl:~~/shaders/Anime4K_Upscale_CNN_x2_S.glsl"; show-text "Anime4K: Mode C (Fast)"
    CTRL+4 no-osd change-list glsl-shaders set "~~/shaders/Anime4K_Clamp_Highlights.glsl:~~/shaders/Anime4K_Restore_CNN_M.glsl:~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl:~~/shaders/Anime4K_Restore_CNN_S.glsl:~~/shaders/Anime4K_AutoDownscalePre_x2.glsl:~~/shaders/Anime4K_AutoDownscalePre_x4.glsl:~~/shaders/Anime4K_Upscale_CNN_x2_S.glsl"; show-text "Anime4K: Mode A+A (Fast)"
    CTRL+5 no-osd change-list glsl-shaders set "~~/shaders/Anime4K_Clamp_Highlights.glsl:~~/shaders/Anime4K_Restore_CNN_Soft_M.glsl:~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl:~~/shaders/Anime4K_AutoDownscalePre_x2.glsl:~~/shaders/Anime4K_AutoDownscalePre_x4.glsl:~~/shaders/Anime4K_Restore_CNN_Soft_S.glsl:~~/shaders/Anime4K_Upscale_CNN_x2_S.glsl"; show-text "Anime4K: Mode B+B (Fast)"
    CTRL+6 no-osd change-list glsl-shaders set "~~/shaders/Anime4K_Clamp_Highlights.glsl:~~/shaders/Anime4K_Upscale_Denoise_CNN_x2_M.glsl:~~/shaders/Anime4K_AutoDownscalePre_x2.glsl:~~/shaders/Anime4K_AutoDownscalePre_x4.glsl:~~/shaders/Anime4K_Restore_CNN_S.glsl:~~/shaders/Anime4K_Upscale_CNN_x2_S.glsl"; show-text "Anime4K: Mode C+A (Fast)"
    CTRL+0 no-osd change-list glsl-shaders clr""; show-text "GLSL shaders cleared""
    

    GLSL files in the shaders directory image

    Output of Shift+I 2 on mpv image

    Desktop (please complete the following information):

    • OS: Ultramarine Linux(Fedora based distribution) 64 Bit
    • Version: v4.0
    • GPU: Nvidia GTX 1650
    • Driver version: 495.46(X11)
    • Media Player and Version: mpv 0.34.1
    • Browser: Not applicable

    Additional context Add any other context about the problem here.

    opened by VihagChaturvedi 8
  • Noob here, Can someone giude me to run this?

    Noob here, Can someone giude me to run this?

    Hi,

    Noob here so please avoid my ignorance. I am not well versed with the field but would like to convert &/or view Anime. Even to 1080p of old once from 480p will be awesome!

    I own a 2017 macPro with Raedon 560.

    Thanks!

    opened by pranavburnwal 8
  • GLSL Version for OpenGL based players (i.e. mpv) and Linux

    GLSL Version for OpenGL based players (i.e. mpv) and Linux

    Hey,

    I think it would be nice to have GLSL versions of this, especially since as far as I'm aware, a large portion of the more video-tweaky community often use's Linux and OpenGL based players like mpv, rather than DirectX.

    enhancement 
    opened by Waiifu 8
  • Insane FPS lag

    Insane FPS lag

    As soon as I try to use the A+A(HQ) in mpv while watching a 1080p anime video, my video is only playing around 3 frames every 3 seconds, though the audio is fine. : CTRL+5 no-osd change-list glsl-shaders set "~~/shaders/Anime4K_Clamp_Highlights.glsl;~~/shaders/Anime4K_Restore_CNN_VL.glsl;~~/shaders/Anime4K_Upscale_CNN_x2_VL.glsl;~~/shaders/Anime4K_Restore_CNN_M.glsl;~~/shaders/Anime4K_AutoDownscalePre_x2.glsl;~~/shaders/Anime4K_AutoDownscalePre_x4.glsl;~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl"; show-text "Anime4K: Mode A+A (HQ)"

    I tried switching out some of the shaders to see which one was causing the issue. It turns out Anime4K_Upscale_CNN_x2_VL.glsl is the one giving me the trouble. If I switched to: Anime4K_Upscale_CNN_x2_M.glsl the problem goes away.

    I noticed that the problem only presents itself if I tried to re-size or full-screen the video. If I just left the window size alone, the preset worked flawlessly.

    To Reproduce Steps to reproduce the behavior:

    1. Open a 1920x1080p anime video.
    2. Full-screen the video
    3. Enable the A+A(HQ) preset

    Expected behavior Video frames start to insanely lag, producing only a couple of frames every few seconds. The audio remains unchanged.

    Screenshots If applicable, add screenshots to help explain your problem.

    Desktop (please complete the following information):

    • OS: Windows 10 -Monitor Resolution: 3440x1440
    • Version: 10.0.19043 Build 19043
    • GPU: RTX 2080 Super
    • Media Player and Version: MPV v0.33
    • Browser (If applicable) [e.g. chrome, safari]

    Additional context My mpv.conf: `#

    THIS FILE WILL BE OVERWRITTEN WITH THE PACKAGE UPDATE

    Use %APPDATA%/mpv/mpv.conf

    OR SVP's menu -> Information -> Add. information -> mpv config. file

    to define your own customized configuration

    Sample mpv configuration

    this is required for SVP to "catch" the mpv

    input-ipc-server=mpvpipe

    hardware video decoder

    hwdec=auto-copy hwdec-codecs=all

    sub-font-size=40

    increase subtitle font size

    ALT+k add sub-scale +0.1

    decrease subtitle font size

    ALT+j add sub-scale -0.1

    high quality video output, require rather fast video card

    D3D11 renderer (default) is required for the HDR playback

    profile=gpu-hq gpu-api=d3d11 scale=ewa_lanczossharp cscale=ewa_lanczossharp video-sync=display-resample

    miscellaneous options

    comment this one if you want mpv to remember the playback position

    save-position-on-quit=no

    fixes audio desync

    hr-seek-framedrop=no

    "ReClock" replacement

    #video-sync=display-resample #video-sync-max-video-change=5

    #log-file=mpv.log

    increase subtitle font size

    ALT+k add sub-scale +0.1

    decrease subtitle font size

    ALT+j add sub-scale -0.1 `

    opened by pufss 7
  • Improved Instructions for Linux and Minor change to Issue Template

    Improved Instructions for Linux and Minor change to Issue Template

    Hi, sorry for the late PR :P

    Haven't had much time recently so I just managed to squeeze this in

    Do need a little bit of revamp for Windows. Won't mind doing it if requested

    Alternatively, this can be merged into the wiki as i believe it makes much more sense to put it in there

    Also, I have added a small change to the Issue Template for bugs to give more details for Linux users, just to make it a bit more easier to troubleshoot

    opened by VihagChaturvedi 6
  • 4.0.1 Release, but asset zip is 4.0

    4.0.1 Release, but asset zip is 4.0

    Describe the bug This is release link of 4.0.1:
    https://github.com/bloc97/Anime4K/releases/tag/v4.0.1

    The assets area have a zip file called Anime4K_v4.0.zip
    This zip file got SHA256 139CD282086457C5ADC79CAF7B75B8B825091D71C9B54958C18745FEA62D7ED7
    It's a same file to early 4.0 version, not new 4.1.

    To Reproduce Steps to reproduce the behavior:

    1. Go to 'https://github.com/bloc97/Anime4K/releases/tag/v4.0.1'
    2. Click on 'Anime4K_v4.0.zip'

    Expected behavior Release archive for 4.0.1 version.

    Screenshots

    Desktop (please complete the following information):

    • OS: e.g. Windows 10
    • Version v4.0.1
    • GPU e.g. GTX 1080
    • Media Player and Version e.g. MPV
    • Browser e.g. chrome

    Additional context

    opened by mhtvsSFrpHdE 5
  • "Further details about each shader and its purpose will be released soon."

    I know that you are probably busy with other things. But I'm still waiting for details on v3.0 to use it. It's just that it looks like you forgot. Anyway, thanks for your shader, I love it!

    opened by HenDGS 5
  • May I request to get the

    May I request to get the "/hdd/sdb/SYNLA_Plus_4096.npy" which is used in the Train_Model.ipynb?

    Thank you for your excellent work in Anime4K! Now I am trying to learn and train the contents of the "Tensorflow" folder, so may I request to get the "/hdd/sdb/SYNLA_Plus_4096.npy" which is used in the Train_Model.ipynb? I'd appreciate it if you would tell me~

    opened by kimwao 1
  • Question on how I can learn

    Question on how I can learn

    Hi,

    I'm a software dev who uses Scala at work. I would like to learn more about AI and deep learning and was amazed by your work here. I was hoping you could give me pointers on how I could learn this specific field of AI better to hopefully make useful software for animators and artists one day.

    How much theory should I go and learn? I learned programming by lots of self projects. I enjoy reading research papers. Would you recommend going through with them and experimenting?

    Much thanks!

    opened by danjavady 1
  • Not working on macOS with Apple Silicon

    Not working on macOS with Apple Silicon

    Fortunately I was able to follow this guide (https://atlassc.net/2022/01/05/anime4k-on-m1-max/) and got it to work. The article OP only specifies a single profile but I went ahead and made the equivalent profiles for the modes in the Windows and Linux instructions, as seen below. Feel free to add them to a macOS install guide.

    # mpv.conf
    
    [Anime4K_A_A]
    glsl-shaders-append="~~/shaders/Anime4K_Clamp_Highlights.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_VL.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_VL.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x2.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x4.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl"
    
    [Anime4K_A]
    glsl-shaders-append="~~/shaders/Anime4K_Clamp_Highlights.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_VL.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_VL.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x2.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x4.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl"
    
    [Anime4K_B_B]
    glsl-shaders-append="~~/shaders/Anime4K_Clamp_Highlights.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_Soft_VL.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_VL.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x2.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x4.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_Soft_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl"
    
    [Anime4K_B]
    glsl-shaders-append="~~/shaders/Anime4K_Clamp_Highlights.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_Soft_VL.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_VL.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x2.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x4.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl"
    
    [Anime4K_C_A]
    glsl-shaders-append="~~/shaders/Anime4K_Clamp_Highlights.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_Denoise_CNN_x2_VL.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x2.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x4.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl"
    
    [Anime4K_C]
    glsl-shaders-append="~~/shaders/Anime4K_Clamp_Highlights.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_Denoise_CNN_x2_VL.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x2.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x4.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl"
    
    [Anime4K_Fast_A_A]
    glsl-shaders-append="~~/shaders/Anime4K_Clamp_Highlights.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_S.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x2.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x4.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_S.glsl"
    
    [Anime4K_Fast_A]
    glsl-shaders-append="~~/shaders/Anime4K_Clamp_Highlights.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x2.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x4.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_S.glsl"
    
    [Anime4K_Fast_B_B]
    glsl-shaders-append="~~/shaders/Anime4K_Clamp_Highlights.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_Soft_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x2.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x4.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_Soft_S.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_S.glsl"
    
    [Anime4K_Fast_B]
    glsl-shaders-append="~~/shaders/Anime4K_Clamp_Highlights.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_Soft_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x2.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x4.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_S.glsl"
    
    [Anime4K_Fast_C_A]
    glsl-shaders-append="~~/shaders/Anime4K_Clamp_Highlights.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_Denoise_CNN_x2_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x2.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x4.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Restore_CNN_S.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_S.glsl"
    
    [Anime4K_Fast_C]
    glsl-shaders-append="~~/shaders/Anime4K_Clamp_Highlights.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_Denoise_CNN_x2_M.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x2.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_AutoDownscalePre_x4.glsl"
    glsl-shaders-append="~~/shaders/Anime4K_Upscale_CNN_x2_S.glsl"
    
    # input.conf
    
    # High end variants
    CTRL+1 no-osd change-list glsl-shaders clr ""; apply-profile "Anime4K_A_A"; show-text "Profile: Anime4K_A_A"
    CTRL+2 no-osd change-list glsl-shaders clr ""; apply-profile "Anime4K_A"; show-text "Profile: Anime4K_A"
    CTRL+3 no-osd change-list glsl-shaders clr ""; apply-profile "Anime4K_B_B"; show-text "Profile: Anime4K_B_B"
    CTRL+4 no-osd change-list glsl-shaders clr ""; apply-profile "Anime4K_B"; show-text "Profile: Anime4K_B"
    CTRL+5 no-osd change-list glsl-shaders clr ""; apply-profile "Anime4K_C_A"; show-text "Profile: Anime4K_C_A"
    CTRL+6 no-osd change-list glsl-shaders clr ""; apply-profile "Anime4K_C"; show-text "Profile: Anime4K_C"
    
    # Fast variants
    CTRL+1 no-osd change-list glsl-shaders clr ""; apply-profile "Anime4K_Fast_A_A"; show-text "Profile: Anime4K_Fast_A_A"
    CTRL+2 no-osd change-list glsl-shaders clr ""; apply-profile "Anime4K_Fast_A"; show-text "Profile: Anime4K_Fast_A"
    CTRL+3 no-osd change-list glsl-shaders clr ""; apply-profile "Anime4K_Fast_B_B"; show-text "Profile: Anime4K_Fast_B_B"
    CTRL+4 no-osd change-list glsl-shaders clr ""; apply-profile "Anime4K_Fast_B"; show-text "Profile: Anime4K_Fast_B"
    CTRL+5 no-osd change-list glsl-shaders clr ""; apply-profile "Anime4K_Fast_C_A"; show-text "Profile: Anime4K_Fast_C_A"
    CTRL+6 no-osd change-list glsl-shaders clr ""; apply-profile "Anime4K_Fast_C"; show-text "Profile: Anime4K_Fast_C"
    
    opened by satvikpendem 3
  • How to use it in Android GLSurfaceView

    How to use it in Android GLSurfaceView

    I'd like to render image by GLSurfaceView and make the image clearer, but I don't know how to use the anime4k related glsl files. Do you have any suggestions?

    opened by wlxyhy 0
  • Added Flatpak guide and some further changes to guides.md

    Added Flatpak guide and some further changes to guides.md

    1. Updated fedora workstation/spins guide : As mpv is not available in the main official fedora repos, we've to add RPM-Fusion repos.
    2. Added Flatpak guide.
    3. Added new version(latest) screenshot in "Usage Instructions for Anime4K section"
    4. Added a note and a tip for fedora and flatpak guide.
    opened by abhigit23 7
  • Can anime4K enlarge the video and save it?

    Can anime4K enlarge the video and save it?

    Is your feature request related to a problem? Please describe. A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]

    Describe the solution you'd like A clear and concise description of what you want to happen.

    Describe alternatives you've considered A clear and concise description of any alternative solutions or features you've considered.

    Additional context Add any other context or screenshots about the feature request here.

    opened by xiaowu2001 10
Releases(v4.0.1)
  • v4.0.1(Sep 17, 2021)

    Installation Instructions for GLSL/MPV (v4.0.1)

    What's new:

    • High quality Restore_Soft shaders (5 variants) to address different types of artifacts. Soft variants significantly reduces ringing, aliasing and other artifacts at the expense of less sharpness.
    • Simplified default shaders usage without compromising on customizability.
    • Complicated use cases (>12 different shader combinations) were simplified and grouped into 3 modes (A, B and C). (See installation instructions)

    Major Changes:

    • Better installation/customization guide.
    • Light variants of Restore shaders are replaced by Soft variants.
    • Improved line restoration shaders.

    Minor Changes:

    • Restore_Moderate shaders are renamed to simply Restore.
    • Added AutoDownscalePre_x2 shader for better compatibility with 1440p/1080p monitors.
    Source code(tar.gz)
    Source code(zip)
    Anime4K_v4.0.zip(758.10 KB)
  • v4.0.0-RC(Aug 10, 2021)

    Installation Instructions for GLSL/MPV (v4.0)

    What's new:

    • Every change from v3.2 is included.
    • Ten variants for line reconstruction shaders. 2 Strengths with 5 network sizes (S/M/L/VL/UL).
    • Improved line reconstruction shaders.
    • Reduced unwanted ringing, aliasing and banding artifacts for line reconstruction shaders.
    • Improved text legibility for line reconstruction shaders.

    Major Changes:

    • All shaders now work on RGB space instead of Luma.
    • This version breaks compatibility with all versions before v3.2! Mixing with older shaders might cause issues.
    • Removed RA, Upscale+Deblur CNN and Deblur CNN shaders, as they are superseded by v4.x Restore shaders.
    • Removed Heavy variant for line reconstruction. The Moderate variant has been tweaked for better reconstruction of low quality images.

    Minor Changes:

    • Standardized network/model size convention. Each increase in size doubles the amount of computation required.
    • Standardized shader naming convention. All shader names now take the form Anime4K-[Type-Subtype]-[Strength]-[Quality/Speed].
    • Line reconstruction shaders renamed to a shorter name: Restore.
    Source code(tar.gz)
    Source code(zip)
    Anime4K_v4.0.zip(756.33 KB)
  • v3.2(Aug 10, 2021)

    What's new:

    • Improved CNN upscale and denoise shaders.
    • Ten variants for upscale shaders. 2 types (None/Denoise) with 5 network sizes (S/M/L/VL/UL).

    Major Changes:

    • Updated shaders for compatibility with future v4.x shaders.
    • All shaders now work on RGB space instead of Luma.
    • This version breaks compatibility with all previous versions! Mixing with older shaders might cause issues.
    • Removed RA, Upscale+Deblur CNN and Deblur CNN shaders, as they are superseded by v4.x Restore shaders.

    Minor Changes:

    • Standardized network/model size convention. Each increase in size doubles the amount of computation required.
    • Standardized shader naming convention. All shader names now take the form Anime4K-[Type-Subtype]-[Strength]-[Quality/Speed].
    • Non-CNN Upscale and Deblur shaders are now deprecated. They might be removed in future versions.
    • Small tweaks for non-CNN shaders improving quality for low resolution inputs.
    • Renamed DarkLines and ThinLines shaders to a shorter name: Darken and Thin.
    Source code(tar.gz)
    Source code(zip)
    Anime4K_v3.2.zip(388.14 KB)
  • v4.0.0(Jul 26, 2021)

    Huge quality improvement on ITW (In the wild) anime. Previous super resolution methods were suited for artificially downscaled anime, while anime found in the wild exhibit a large amounts of variance. We propose and develop a method to correct this distribution shift for already existing CNN upscalers and that can be applied to non-Anime4K upscalers (The upscaling shaders bundled in this version are the exact same as v3.1). The line reconstruction algorithm will be explained in a future informal paper.

    Some small issues such as occasional ghosting or ringing will need to be fixed before official release.

    The three versions of line reconstruction shaders can be used as follows in mpv's input.conf: CTRL+2 works on almost all 1080p anime, if blurriness and aliasing is noticeable, use the heavier version (CTRL+3), if ghosting and ringing is visible, use the lighter version (CTRL+1). For downscaled 720/480p anime, start with CTRL+1, for native standard definition anime, start with CTRL+2.

    input.conf for 4K screens (screen size, not anime size)

    CTRL+1 change-list glsl-shaders set "~~/shaders/Anime4K_Line_Reconstruction_Light_L.glsl;~~/shaders/Anime4K_Upscale_CNN_L_x2_Denoise.glsl;~~/shaders/Anime4K_Auto_Downscale_Pre_x4.glsl;~~/shaders/Anime4K_Line_Reconstruction_Heavy_L.glsl;~~/shaders/Anime4K_Upscale_CNN_M_x2_Denoise.glsl;~~/shaders/Anime4K_DeRing.glsl;"
    CTRL+2 change-list glsl-shaders set "~~/shaders/Anime4K_Line_Reconstruction_Medium_L.glsl;~~/shaders/Anime4K_Upscale_CNN_L_x2_Denoise.glsl;~~/shaders/Anime4K_Auto_Downscale_Pre_x4.glsl;~~/shaders/Anime4K_Line_Reconstruction_Heavy_L.glsl;~~/shaders/Anime4K_Upscale_CNN_M_x2_Denoise.glsl;~~/shaders/Anime4K_DeRing.glsl;"
    CTRL+3 change-list glsl-shaders set "~~/shaders/Anime4K_Line_Reconstruction_Heavy_L.glsl;~~/shaders/Anime4K_Upscale_CNN_L_x2_Denoise.glsl;~~/shaders/Anime4K_Auto_Downscale_Pre_x4.glsl;~~/shaders/Anime4K_Line_Reconstruction_Medium_L.glsl;~~/shaders/Anime4K_Upscale_CNN_M_x2_Denoise.glsl;~~/shaders/Anime4K_DeRing.glsl;"
    CTRL+0 change-list glsl-shaders clr ""
    

    input.conf for 1080p screens (screen size, not anime size)

    CTRL+1 change-list glsl-shaders set "~~/shaders/Anime4K_Line_Reconstruction_Light_L.glsl;~~/shaders/Anime4K_Upscale_CNN_L_x2_Denoise.glsl;~~/shaders/Anime4K_Auto_Downscale_Pre_x4.glsl;~~/shaders/Anime4K_Line_Reconstruction_Medium_L.glsl;~~/shaders/Anime4K_Upscale_CNN_M_x2_Denoise.glsl;~~/shaders/Anime4K_DeRing.glsl;"
    CTRL+2 change-list glsl-shaders set "~~/shaders/Anime4K_Line_Reconstruction_Medium_L.glsl;~~/shaders/Anime4K_Upscale_CNN_L_x2_Denoise.glsl;~~/shaders/Anime4K_Auto_Downscale_Pre_x4.glsl;~~/shaders/Anime4K_Line_Reconstruction_Light_L.glsl;~~/shaders/Anime4K_Upscale_CNN_M_x2_Denoise.glsl;~~/shaders/Anime4K_DeRing.glsl;"
    CTRL+3 change-list glsl-shaders set "~~/shaders/Anime4K_Line_Reconstruction_Heavy_L.glsl;~~/shaders/Anime4K_Upscale_CNN_L_x2_Denoise.glsl;~~/shaders/Anime4K_Auto_Downscale_Pre_x4.glsl;~~/shaders/Anime4K_Line_Reconstruction_Light_L.glsl;~~/shaders/Anime4K_Upscale_CNN_M_x2_Denoise.glsl;~~/shaders/Anime4K_DeRing.glsl;"
    CTRL+0 change-list glsl-shaders clr ""
    

    Note: This version is only compatible with v3.1 and older shaders. Mixing with newer shaders might cause issues.

    Source code(tar.gz)
    Source code(zip)
    Anime4K_v4.0_Experimental.zip(67.13 KB)
  • 3.1(May 30, 2020)

    Installation Instructions for GLSL/MPV (v3.1)

    What's new:

    Minor changes:

    • Removed the version number from the file name. Users will not have to update the config file anymore for future updates.

    Bug fixes:

    • Fixed texture size issues in RA, Upscale_CNN, and Upscale_CNN_Deblur shaders that caused faulty processing when stacked one after the other.
    • RA shaders now correctly ignore chroma channels. They used to wrongly process chroma channels.
    Source code(tar.gz)
    Source code(zip)
    Anime4K_v3.1.zip(612.51 KB)
  • v3.0(Aug 10, 2021)

    Installation Instructions for GLSL/MPV (v3.0)

    What's new:

    • The shader is now broken into modular parts, allowing extensive customization.
    • Better quality CNN line art super-resolution shaders. (Faster and higher PSNR compared to all NGU and FSRCNNX variants.)
    • A large amount of shaders designed to improve visual/perceptual quality.
    • Shaders designed to reduce noise and artifacts.

    Minor changes:

    • The SR CNNs were trained on MAE loss with the SYNLA dataset, instead of perceptual loss and images randomly collected in the wild in v2.1. This significantly improves accuracy for recovering detail in low resolution videos at the cost of some perceptual quality. However this can be mitigated by using the aforementioned shaders.
    Source code(tar.gz)
    Source code(zip)
    Anime4K_v3.0.zip(606.75 KB)
  • v2.1a(Mar 26, 2020)

    Installation Instructions for GLSL/MPV Version

    Hotfix: Algorithm now performs downscaling before applying the x4 prescaler (rather than after). 720p scaled to x4 is bigger than 4K, which is unnecessary and slows the algorithm down.

    Added a L version that contains a larger prescaler. It offers a significant improvement to image quality, especially to low resolution anime. (eg. 480p) It has 2741 parameters, which is still smaller than FSRCNNX 8. If your GPU can run FSRCNNX, it should be fine. However, since the prescaler was trained using perceptual loss, some banding might occur on very old/blurry anime. It is also slower. (+1ms compared to non-L version for 1080p -> 4K upscaling) Use at your own discretion.

    Source code(tar.gz)
    Source code(zip)
    Anime4K_Hybrid_v2.1a.zip(50.82 KB)
  • v2.1(Mar 22, 2020)

    Installation Instructions for GLSL/MPV Version

    Integrated a pre-scaler in Anime4K. It is a tiny ConvNet (916 parameters, 3 hidden layers of 4 feature maps) trained with the differentiable Anime4K algorithm and other techniques that will be published soon. For comparison, FSRCNNX has more than 3000 parameters (depending on the version) and waifu2x has more than 1 million.

    This pre-scaler gives a major quality improvement to lower resolution anime, especially for 720p and 480p. 480p -> 1080p, 480p -> 4K and 720p -> 4K are now officially supported. Texture loss and aliasing issues are again reduced.

    The pre-scaler uses 1.5ms for 1080p -> 4K upscaling, bringing the total to 4ms on a Vega 64 GPU.

    Source code(tar.gz)
    Source code(zip)
    Anime4K_Hybrid_v2.1.zip(10.70 KB)
  • v2.0-RC5(Mar 11, 2020)

    Installation Instructions for GLSL/MPV Version

    Replaced the Bilateral Mean Filter with a "Regularized" Bilateral Mode Filter. The mode is more robust than the mean or median, and introduces the least amount of blur.

    The mode filter needs to be "regularized" since the mode of a set is not defined when the number of data points is finite and contains unique data (all data points are different). This undefined case happens extremely often in images when using a small kernel (eg. 3x3), since there are 256 possible luma values, but only 9 samples. Often all 9 pixels will have different values, and rigging up a continuous fallback to median/mean that does not introduce discontinuities is not trivial.

    However, we can use Kernel Density Estimation (KDE) as a continuous approximation of the discrete probability color histogram.

    Source code(tar.gz)
    Source code(zip)
    Anime4K_Hybrid_v2.0RC5.zip(3.46 KB)
  • v2.0-RC4(Mar 10, 2020)

    Installation Instructions for GLSL/MPV Version

    Added a bilateral filter to Anime4K. This reduces noise and blocking on highly compressed video (mostly found on the web). Speed is not affected.

    Anime4K now enhances edges even in cases where upscaling is not necessary, as seen in this comparison.

    On a more personal note, I now feel as if ML based techniques for denoising line art are just mostly re-learning traditional computer graphics/image processing techniques, + ε. The small epsilon leads to a slightly better image quality, but at what cost to speed? (Of course this does not apply for extreme cases such as low resolution upscaling, denosing with a negative signal to noise ratio or image style transfer, since ConvNets can "generate" missing information).

    image

    Source code(tar.gz)
    Source code(zip)
    Anime4K_Hybrid_v2.0RC4.zip(3.63 KB)
  • v2.0-RC3(Mar 3, 2020)

  • v2.0-RC2(Feb 29, 2020)

    Installation Instructions for GLSL/MPV Version

    Further speed improvements without noticeable quality degradation. Three versions are provided:

    • Normal: 2.5ms
    • Fast: 1ms (Texture detector removed + Low resolution gradient computation)
    • UltraFast: 0.8ms (Crude gradient computation)

    All tests are carried out using a Vega 64 GPU using default settings, with the MPV profiler (accessed by pressing Shift+I, then pressing 2).

    Note that Fast and UltraFast versions might still produce aliasing artifacts since all gradient computation is done at the original video size (before upscaling).

    Source code(tar.gz)
    Source code(zip)
    Anime4K_Hybrid_v2.0RC2.zip(9.76 KB)
  • v2.0-RC(Feb 29, 2020)

    Installation Instructions for GLSL/MPV Version

    Major changes to the algorithm. It is now a completely continuous algorithm, discrete comparison kernels are removed. A useful side effect is that the refine step is now fully differentiable (future unification with Maching Learning!)

    Major quality improvements without speed penalties. Texture loss and aliasing issues are solved. Improved 720p->1080p and 1080p->1440p upscaling quality. Due to it being completely continuous, it now also works with non-anime (real) videos such as movies, games and youtube videos.

    Further details (algorithm details, explanation, examples and preprint) will be drafted soon.

    Source code(tar.gz)
    Source code(zip)
    Anime4K_Hybrid_v2.0RC.zip(3.64 KB)
  • v1.0-RC2(Oct 10, 2019)

  • v1.0-RC(Aug 19, 2019)

    Installation Instructions for GLSL/MPV Version

    Reduced texture loss, aliasing and banding in Anime4K v1.0 RC at the cost of performance. It now takes 6ms. +2ms for line detection and +1ms for line targeted FXAA.

    What's new:

    • A line detection algorithm.
    • Gradient maximization is only applied near lines using the line detector, instead of indiscriminately affecting the entire image. This has the effect of ignoring textures and out of focus elements.
    • Finally, one iteration of targeted FXAA is applied on the lines using the line detector to reduce aliasing.

    ComparisonRC

    Source code(tar.gz)
    Source code(zip)
    Anime4K_GLSL_v1.0RC.zip(3.06 KB)
  • v0.9(Aug 14, 2019)

Owner
CS Student in Machine Learning, Image Processing, Computer Graphics and Scientific Simulation
Visyerres sgdf woob - Modules Woob pour l'intranet et autres sites Scouts et Guides de France

Vis'Yerres SGDF - Modules Woob Vous avez le sentiment que l'intranet des Scouts

Thomas Touhey (pas un pseudonyme) 3 Dec 24, 2022
Code for the prototype tool in our paper "CoProtector: Protect Open-Source Code against Unauthorized Training Usage with Data Poisoning".

CoProtector Code for the prototype tool in our paper "CoProtector: Protect Open-Source Code against Unauthorized Training Usage with Data Poisoning".

Zhensu Sun 1 Oct 26, 2021
Similarity-based Gray-box Adversarial Attack Against Deep Face Recognition

Similarity-based Gray-box Adversarial Attack Against Deep Face Recognition Introduction Run attack: SGADV.py Objective function: foolbox/attacks/gradi

1 Jul 18, 2022
Learning Representational Invariances for Data-Efficient Action Recognition

Learning Representational Invariances for Data-Efficient Action Recognition Official PyTorch implementation for Learning Representational Invariances

Virginia Tech Vision and Learning Lab 27 Nov 22, 2022
Rayvens makes it possible for data scientists to access hundreds of data services within Ray with little effort.

Rayvens augments Ray with events. With Rayvens, Ray applications can subscribe to event streams, process and produce events. Rayvens leverages Apache

CodeFlare 32 Dec 25, 2022
Unified API to facilitate usage of pre-trained "perceptor" models, a la CLIP

mmc installation git clone https://github.com/dmarx/Multi-Modal-Comparators cd 'Multi-Modal-Comparators' pip install poetry poetry build pip install d

David Marx 37 Nov 25, 2022
GAN encoders in PyTorch that could match PGGAN, StyleGAN v1/v2, and BigGAN. Code also integrates the implementation of these GANs.

MTV-TSA: Adaptable GAN Encoders for Image Reconstruction via Multi-type Latent Vectors with Two-scale Attentions. This is the official code release fo

owl 37 Dec 24, 2022
A generator of point clouds dataset for PyPipes.

CloudPipesGenerator Documentation | Colab Notebooks | Video Tutorials | Master Degree website A generator of point clouds dataset for PyPipes. TODO Us

1 Jan 13, 2022
Code for ICLR 2020 paper "VL-BERT: Pre-training of Generic Visual-Linguistic Representations".

VL-BERT By Weijie Su, Xizhou Zhu, Yue Cao, Bin Li, Lewei Lu, Furu Wei, Jifeng Dai. This repository is an official implementation of the paper VL-BERT:

Weijie Su 698 Dec 18, 2022
Tool which allow you to detect and translate text.

Text detection and recognition This repository contains tool which allow to detect region with text and translate it one by one. Description Two pretr

Damian Panek 176 Nov 28, 2022
Visual Tracking by TridenAlign and Context Embedding

Visual Tracking by TridentAlign and Context Embedding (TACT) Test code for "Visual Tracking by TridentAlign and Context Embedding" Janghoon Choi, Juns

Janghoon Choi 32 Aug 25, 2021
In this work, we will implement some basic but important algorithm of machine learning step by step.

WoRkS continued English 中文 Français Probability Density Estimation-Non-Parametric Methods(概率密度估计-非参数方法) 1. Kernel / k-Nearest Neighborhood Density Est

liziyu0104 1 Dec 30, 2021
Framework for estimating the structures and parameters of Bayesian networks (DAGs) at per-sample resolution

Sample-specific Bayesian Networks A framework for estimating the structures and parameters of Bayesian networks (DAGs) at per-sample or per-patient re

Caleb Ellington 1 Sep 23, 2022
⚖️🔁🔮🕵️‍♂️🦹🖼️ Code for *Measuring the Contribution of Multiple Model Representations in Detecting Adversarial Instances* paper.

Measuring the Contribution of Multiple Model Representations in Detecting Adversarial Instances This repository contains the code for Measuring the Co

Daniel Steinberg 0 Nov 06, 2022
Local Similarity Pattern and Cost Self-Reassembling for Deep Stereo Matching Networks

Local Similarity Pattern and Cost Self-Reassembling for Deep Stereo Matching Networks Contributions A novel pairwise feature LSP to extract structural

31 Dec 06, 2022
CvT-ASSD: Convolutional vision-Transformerbased Attentive Single Shot MultiBox Detector (ICTAI 2021 CCF-C 会议)The 33rd IEEE International Conference on Tools with Artificial Intelligence

CvT-ASSD including extra CvT, CvT-SSD, VGG-ASSD models original-code-website: https://github.com/albert-jin/CvT-SSD new-code-website: https://github.c

金伟强 -上海大学人工智能小渣渣~ 5 Mar 07, 2022
Code in PyTorch for the convex combination linear IAF and the Householder Flow, J.M. Tomczak & M. Welling

VAE with Volume-Preserving Flows This is a PyTorch implementation of two volume-preserving flows as described in the following papers: Tomczak, J. M.,

Jakub Tomczak 87 Dec 26, 2022
Kaggle Lyft Motion Prediction for Autonomous Vehicles 4th place solution

Lyft Motion Prediction for Autonomous Vehicles Code for the 4th place solution of Lyft Motion Prediction for Autonomous Vehicles on Kaggle. Discussion

44 Jun 27, 2022
Barbershop: GAN-based Image Compositing using Segmentation Masks (SIGGRAPH Asia 2021)

Barbershop: GAN-based Image Compositing using Segmentation Masks Barbershop: GAN-based Image Compositing using Segmentation Masks Peihao Zhu, Rameen A

Peihao Zhu 928 Dec 30, 2022
Code repo for "RBSRICNN: Raw Burst Super-Resolution through Iterative Convolutional Neural Network" (Machine Learning and the Physical Sciences workshop in NeurIPS 2021).

RBSRICNN: Raw Burst Super-Resolution through Iterative Convolutional Neural Network An official PyTorch implementation of the RBSRICNN network as desc

Rao Muhammad Umer 6 Nov 14, 2022