Personal View site logo
Cinema gear deals, direct from factories - Gear deals and Gear deals section. Also check Cameras, lenses, software, gear deals.
You support is vital for us. To keep this place ad free and independent, select one of the options below.
Donations are going to laboratory, support costs, hosting, etc. Your support allows to improve and expand this site.
VideoArtifact: Footage Quality Restoration Software
  • Video Artifact is a new standalone video editor for digital film restoration and footage quality enhancement.

    It has free and paid editions.

    http://www.videoartifact.com/va/

    Program helps you to prepare footages from consumer or professional video cameras, digicams and V-DSLR.

    It can remove MJPEG, MPEG, H.264, H.265 compression artifacts and digital noise, improves sharpness and color resolution up to 16 bit 4:4:4.

    Restoration features:

    • Script-based to work with unique cases. You can use many filters in any combination.
    • 50+ specialized restoration filters work in Y, U, V separately in 8/16 bit.
    • Possibility to use dozens of different type denoisers for different noise type.
    • Accurate denoising that keeps micro details with no "plastic" effect.
    • Removing of sharping halos.
    • Linear color high quality spline-based resize to resize up to 4K. Resize halos can be removed.
    • Precise halo-free radius-dependent sharping.
    • Film grain emulation and 3D LUT grading.
    • 4:2:0 8 bit to 4:2:2/4:4:4 16 bit chroma reconstruction.

    It can be used in a post-production workflow if you use V-DSLR or consumer cameras.
    Currently 50+ camera models supported and hundreds of compatible models.

    Special features:

    • Batch processing without GPU. Very fast: 5 fps for typical V-DSLR AVCHD source and 2 fps for 4K.
    • Automatic joining of MTS, MOV and MP4 spans and renaming to yyyymmdd_hhmmss.mov format.
    • Non-standard 16-255 to TV 16-235 levels conversion.
    • ITU-R BT.601 ↔ BT.709 floating point conversion.
    • NLE and DaVinci Resolve compatible workflow.
    • Mastering to webm VP9/Opus or mp4 H.264/AAC 1080p/720p/360p/240p.

    Examples with various restoration methods are available at http://www.videoartifact.com/examples/

    Some samples:

    image image

    image image

  • 63 Replies sorted by
  • @rean glad to see this coming out. What version of windows will this run on, any chance Windows 7 Professional...?

    Any plans for a Mac version?

  • I don't think there will be a mac version, because it's rely on AviSynth. But it's worth every penny and results are quite astonishing.

  • Thanks much for this, looks very interesting. Up until now I have used Avisyth scripts which were, yes, very slow, but very amazing (QTGMC, for example)... especially for older interlaced material.

    A bit more documentation at your website would help, or is the documentation of filters in the download package itself? I am aware there is no one-recipe solution for all material, but more info would be nice.

    Regarding what system requirements asked earlier, it is on the site, says:

    64-bit Windows 7/8/10 PC, 8 GB RAM and stable CPU cooling is required. 4-core Intel I7 3+ GHz is recommended.

    best, ninetto

  • @CFreak you can test it on a Windows emulator. For permanent and professional use you can make a power Windows-based computer. As I know we can buy 2 fast Windows computers on a price of a powerful Mac-branded PC.

    VideoArtifact does not use GPU, so you can save money additionally. A cheap GPU and a cheap power supply allows you to buy better processor.

    I have two computers for my post-production tasks. This saves a lot of time, because I do all tasks in parallel. For example, I encode and process sources on a secondary PC with a cheap video card while I work with a color grade on Davinci Resolve on a primary PC where a powerful CUDA-based GPU is installed.

    About Windows 7, if you asked about a 32-bit PC, this way is not usable. My AviSynth code eats ~3.5 GB RAM and also encoder requires ~ 2 GB additionally. So a x64 system is required. My code is Windows 7+ compatible.

    @ninetto community-based AviSynth code (like QTGMC or sharping scripts) is very slow for Full HD+ resolutions. All VideoArtifact scripts are pure original speed-optimized code with very little quality loss. For most cases I process Y-channel only. So it is a blazing fast in compare! Most used AviSynth scripts are C++ coded to fast process 4K 4:4:4 in 16-bit. Also I use own very optimized AviSynth MT build.

    AviSynth-lowers can also switch their AviSynth setup to VideoArtifact Core-based to use your own scripts. Some users already thanked me with a donation for my AviSynth build because processing speed. Also VideoArtifact is AviSynth Plus MT compatible!

    A documentation is under development. I will create a blog with detailed texts, screenshots, usage examples and tutorials later.

  • @rean I run a Hackintosh (so I don't pay the Mac tax either) and have it set up with a windows 7 partition for a dual boot system. I am pretty sure it's Win 7 Pro, 64-bit, will have to check, I never use the windows side. Thanks for your response.

  • A YouTube tutorial is published. Russian and English subtitles. VideoArtifact Basic + DaVinci Resolve. Improve Video Quality.

  • Do you plan on making a frontend with sliders and a browser, rather than scripting? Also a video of examples would be nice. Something along the lines of Cinnafilm's dark energy demo videos

  • There isn't that much scripting. You only have to read and delete or add a hashtag. It's easy as using a text editor.

  • @robertGL

    rheinpirat is correct.

    I am a private person, not a large programming firm like Adobe, so cannot create a complex software you ask.

    Making of a slider-based editor that has the same flexible as VA has currently is an over-complicated way. To keep all VA features, it should be as complex software as any node-based one, like Nuke or Fusion. Because only node-based editors have the same level of flexibility as scripts can give.

    But VA is designed to be usable for most users, so you do not need in script programming.

    In most cases we only enable or disable something using #, because all required filters already placed on the processing templates, also with universal default parameters. VA has also a personal support of different camera types: DV, MJPEG, MPEG, H.264, 4K H.265, with specific artifact types. Currently more than 55 cameras are fully supported. Also a lot of compatible models can be used, because most hardware cameras have only minor changes like another body, lens, WiFi or something related.

    Some edit things like color correction in VA is not very comfortable, but VA is not for these cases. We can make some color corrections inside the program, but better way is to use specialized software like DaVinci Resolve where it is very easy and fast.

    Please note that VA is optimized to process videos in batch mode. In most cases you will create one-two personalized templates per used camera settings and scene types. Then you will use them in all your 1000 files many months for different projects. For example, I have 2 templates for my NEX-7. I used them with minimal changes in every video project. Most operations are just mouse clicks with very rare text editing. Only rare cases require manual edit, for example, a complex noise in very dark scenes or operator mistakes.

    About examples, yes, I will create a lot of tutorial videos also articles. VA is my the most complex project ever. I spend many-many months of developing, so I do not want to loose it. I am glad to facilitate to beginners understanding of the program. It is normal that something is not fully understand, because the program is very powerful and has many features.

  • Hi @rean, remember the conceptual frontend I developed for your previous software? I could try to do something similar for this. It looks like it would be easier to implement based on how you structured this.

  • Id love to see some more side by side close up before/after examples here.

  • Thanks @rean - I am looking forward to this being successful !

  • I've been trying out the free version. Any chance users will be able to finetune denoise settings with a difference viewer (like in Windmotion). Sharpen possible too?

  • @fancydancer then you will get a "lite" premium edition ;)

    You can only see channels by edit this line:

    global OutView="" # y,u,v,lsb,""

    Sharp is possible by a third-party code. Or just use AviSynth internal 8-bit Sharpen(0.35) before a denoiser. I removed it from a basic template, because bad quality. Paid editions have beautiful 16-bit radius-dependent halo-less sharp with radius 0.01-100 pixels. At this time it is the best AviSynth sharp filter you can find.

    So, why not to order to get features you will love?

  • I have deleted the YouTube video with the moon because found an issue - grain. New version is better. Unfortunately I cannot edit the previous post to change YouTube link.

    http://www.videoartifact.com/examples/va-denoise-moon/

    PS. Video sources from different cameras wanted. I need in licensed camera files to create more different examples. If you have some files for me, please send your files with this information: a copyright string (who is the author), a camera model. You will get additional 50% discount and your camera template support. Please send legal shots. A permission of models is required, and also visible trademarks and another legal issues should not be in the scene.

    Film.png
    975 x 548 - 392K
  • Was just wondering.. How do you want us to provide the videos?
    How many seconds? Do you want multiple samples with different isos? And what about debayered raw - just basic debayer with no additional processing? Tiff sequence or some other file type desired? Thanks

  • @robertGL up to minute is ok with any scene with any camera setting you like to know how good is software.

    Because VA does not support sequenced RAW, you can provide a compressed or not compressed debayered files (avi uncompressed, dnxhd, ProRes or something similar). But if you provide tiff or exr, I will convert them manually by FFmpeg or DaVinci (I use free version only, up to 4K). Also this is a source to make a tutorial how to prepare RAW/sequence files for VA. I can read and write up to 16 bit 4:4:4 mov, mkv or avi, FFmpeg-compatible. The RAW-quality improvement is supported, because VA can make denoise, sharp or quality resize. Some lens/sensor artifacts in edge area are probably can be removed.

    For compressed sources like AVCHD or H.264 you can provide most quality file only. For example, if camera supports 320x200 and 1920x1080, only 1920x1080 is required, because only a few of users will shot at some strange low quality settings. But if you like to get some record modes, you can include.

    Green screen scenes are plus to test edges artifacts.

  • Video Artifact 1.0.0 is just released. Thanks to all users who helped to make this release available. http://www.videoartifact.com/va/

    The Complete VA Filter Reference is also published. 86 filters with around 350 parameters! It was a lot of work to create this document. Now we can understand all filter parameters.

    The next news will be published on the VA homepage. These days I plan to make more manual documents to understand VA better and more examples.

  • Some users ask me about improvement examples from quality sensor cameras. So, I have published a hacked GH2 example processing.

    A youtube preview is also available, but quality is draft.

    The source has only one important issue: color noise. But some other things like sharping or skin tones were improved, too.

    Colors.png
    455 x 256 - 133K
  • @rean: I am somewhat impressed by the sample above. Can you compare with NeatVideo please? Does it yield better results subjectively/objectively on color channel de-noising? I can see a use case in old tape restoration.

  • @radikalfilm I don't know NeatVideo and have no license, so do not know possible trial limitations. But I know users who were impressed after their compare with NeatVideo using VA old versions (called Windmotion) a 1-2 year ago. Probably they are still here on the forum. Try to find them in a Windmotion forum discussion.

    My things as a restoration engineer who work in this area are:

    There is not possible to make objective comparable results. VA has different type denoisers, those can be applied differently for color channels. Also masking and noise shaping can be used to mask or remove digital compression artifacts. In these cases I can create a similar per-channel denoise with NeatVideo, for example, in node-like editor (free DaVinci Resolve can do separate YUV, RGB or HSV processing with NeatVideo) and get similar result, or use it together with NeatVideo over multiple stages. So the result is a combination of many filters and restoration technics knowledge. The important difference will be in the processing speed, learning skills and usability.

    Also color noise removal is not a reliable test. The eye sees very little difference in the color detail. In the examples there is no small color details, so any good denoiser can give you a good result.

    Also applying denoise only is a very specialized test that can show nothing. In VA different denoisers can be used, and it is the correct way to use VA. Probably you would like to test a result for an average user, who do not know special technics and probably will not use advanced features? In this case you can try it yourself with the Basic edition, that has one simple motion-compensate based denoiser but with many parameters for experiments.

    About old tape restoration. VA has no specialized filters for related tape artifacts. But you can use many hundreds of AviSynth community filters, if extend it manually. VA mainly is for digital camera quality and source prepare for film.

  • I suggest making more extreme examples, it's hard for me to get excited about these results, even though your software might be very capable, as the source materials are all quite clean in the first place.

  • @radikalfilm I use NeatVideo on a regular basis and I am also a very avid user of VA since the old version (called Windmotion). NeatVideo is a denoiser with sharpen tools, but VA can simply more. De-Halo, dithering, mosquito noise remove etc. It removes stuff NeatVideo simply can't. Mostly I use VA for preparing my footage for heavy grading, even if the source seems to be clean.

  • I Will try to get some footage for you this weekend from my Nikon V2 and 50D

  • @rheinpirat @rean thank you both for context. @rean since you are a restoration engineer I expect only good things. I even have a specific project in mind, if and when I find the time. I have this short TV film initially shot on PAL DV letterbox, so the raster was only 720x432. I no longer have the source tapes. The final cut has been through denoising+sharpening with NeatVideo, stabilization with Warp Stabilizer (Adobe), 2x upscaling with Image Enhancer (super-resolution software by Infognition), further 1.something upscaling and color correction in Davinci. Davinci correction was on top of a somewhat botched correction in FCP6.

    So now I have this vaguely HD (as in "it fills a 1080p frame") master with color hanging on for dear life (breaking into artifacting if I try to push it more). I can't really think of anything worse to throw at VideoArtifact. I would like to push the colors even more. The master is Cineform 10bit, 4th generation (1st PALDV, 2nd Prores SD, 3rd Cineform HD, 4th Cineform HD) The film is still getting views after all these years. It's aging like good wine.

    Will VA give it a fighting chance for re-remastering in Davinci? Just asking :)