VEAI Command-Line Alpha 3.0.0.2a

Downloads
Windows | Headers, Libs, DLLs
M1 Mac | Headers & Dylibs
Intel Mac | Headers & Dylibs
Source Code (This Release) | Source Code (development branch)

Hello everyone,

The next release of the VEAI 3 alpha is now available! Please note that these initial releases only utilize a subset of FFmpeg features - some features available in general FFmpeg builds may not be available in this build. Source code and redistributables for custom builds are available, so advanced users may be able to enable additional features as desired.

Changelog

  • Memory leak addressed
  • Frame interpolation filter added
  • (Mac) Setup script now adds to .zshrc and .bashrc
  • (Mac) Setup script can better handle some startup scenarios

Windows Installation

  1. Extract the downloaded zip file to a location of your choice
  2. Run the “run-me" file appropriate for your graphics card.
  3. You will be prompted to confirm the auto-detected models location. In most situations, you should be able to just enter “Y”
  4. Follow the on-screen prompts to login to TopazLabs.com and enter your VEAI options
  5. VEAI will run on your files
  6. Once complete you can either exit the window, or type additional commands (see Advanced Usage)

Mac Installation

  1. Extract the downloaded txz file to a location of your choice
  2. Using terminal, run ./installer.command file to set appropriate environment variables.
  3. Run veai-ffmpeg/bin/login email password
  4. Run veai-ffmpeg/run-me.sh input.mp4 to process using Video Toolbox.

Advanced Usage

  • If you have an existing models folder, you may select it by setting the VEAI_MODEL_DIR setting in your system environment variables. If you do not already have downloaded models in this folder, you will need to copy the model JSON files from the download into this folder. It will then be detected by run-me.bat
    • Mac users: this variable is automatically set in .bashrc and .zshrc when you run the setup script, make changes to the existing entries if needed.
  • (Windows) After run-me.bat finishes, you can execute additional VEAI sessions by running FFmpeg. See the end of run-me.bat for an example.
    • As an example: ffmpeg -hide_banner -i "input.mp4" -vf veai_up="scale=2:model=amq-12" "input-2x.mp4" would upscale input by 2x using AMQ 12, saving the output to input-2x.mp4
  • (Mac) The ffmpeg command can be invoked similarly to perform additional VEAI sessions.

Available Scaling Models

  • aaa-10 : Artemis Aliased & Moire v10
  • aaa-9 : Artemis Aliased & Moire v9
  • ahq-11 : Artemis High Quality v11
  • ahq-12 : Artemis High Quality v12
  • alq-12 : Artemis Low Quality v12
  • alq-13 : Artemis Low Quality v13
  • alqs-1 : Artemis Strong Dehalo v1
  • alqs-2 : Artemis Strong Dehalo v2
  • amq-12 : Artemis Medium Quality v12
  • amq-13 : Artemis Medium Quality v13
  • amqs-1 : Artemis Dehalo v1
  • amqs-2 : Artemis Dehalo v2
  • ddv-2 : Dione Interlaced DV v2
  • ddv-3 : Dione Interlaced DV v3
  • dtd-4 : Dione Interlaced Robust v4
  • dtds-2 : Dione Interlaced Robust Dehalo v2
  • dtv-4 : Dione Interlaced TV v4
  • dtvs-2 : Dione Interlaced Dehalo v2
  • gcg-6 : Gaia Computer Graphics v6
  • ghq-6 : Gaia High Quality v6
  • prob-1 : Proteus 6-Parameter v1
  • prob-2 : Proteus 6-Parameter v2
  • thd-3 : Theia Fine Tune Detail v3
  • thd-4 : Theia Fine Tune Detail v4
  • thf-5 : Theia Fine Tune Fidelity v5

Available Interpolation Models

  • chf-1 : Chronos Fast v1
  • chf-2 : Chronos Fast v2
  • chf-3 : Chronos Fast v3
  • chr-1 : Chronos Slo-Mo / FPS Conversion v1
  • chr-2 : Chronos Slo-Mo / FPS Conversion v2

Licenses

  • FFmpeg is licensed under the GNU Lesser General Public License version 2.1 or later (LGPL v2.1+). A copy of this license can be found here .
  • OpenH264 is licensed under the BSD license. A copy can be found here
  • libvpx is licensed under the BSD license. A copy can be found here
8 Likes

Tried Simple 720p video with audio track DD 48khz 5.1 Mkv to AAC 5.1 320kbs Mp4 H265.

everything went well so far.
Had to change the environement table as it was refering to the folder of the previous alpha.

I have tried several times to run this installation unsuccessfully
I have a Windows i9 Pc with a NVidia Geforce RTX 3080 display card.
I have tried to run both the Nvidia run-me and the Intel and I am thrown out when entering my user email and password. I am new to Alpha testing so what am I doing wrong?

Hi, keep getting “Please set the VEAI_MODEL_DIR environment variable to a writable models directory path “”” on my mac.
Folder is writeable and the ~/.bashrc AND ~/.zshrc contain the models path…

Error occurs after install and veai-ffmpeg/bin/login email password

Hi,

Can you try removing the added lines in ~/.bashrc and ~/.zshrc, and then re-download the alpha before running the installer again? I’ve made a small tweak to the installer that will be reflected in a new download.

Hi Kevin,

If you’re comfortable running a command line command through the following steps, it may provide some more detailed output that’s helpful:

  • Shift-right-click an empty area in the same folder as your run-me files, choose Open PowerShell window here.
  • Enter the command .\run-me-nvidia.bat. The script will proceed as normal.

The window should remain open this time, and you’ll be able to see any error messages produced.

It worked once, but it took the models folder from a previous install, although the bashrc and zshrc files looked ok.
When I processed it gave an authentication error, which i can’t reproduce anymore…

The login auth file is stored in the models directory, so if you are using the older models directory then the older auth file will just work fine.

You will need to copy over the chf and Chr files from the new models directory for Chronos to work with the veai_fi filter.

OK, seems to work, thanks

Thanks for the update.

Is this new?

Good catch, no there are no model updates.
If you have a model/version missing and need it then add it to the thread here.

Like the original Gaia CG.

I just tried piping output from Vapoursynth (QTGMC deinterlace / denoise / derainbow, downscale & a few others) directly into veai-ffmpeg to upscale with Proteus, all in one command-line operation.

Worked a treat. This is getting exciting.

2 Likes

I tried the following scaling models successfully with run-me-nvidia.bat:
prob-2 / up=1 / q=10 ==> fps=0.4
thd-4 / up=1 / q=24 ==> fps=0.8
thd-4 / up=4 / q=24 ==> fps=0.4
ghq-6 / up=4 / q=24 ==> fps=0.2

thf-5 did not find parameters and returned without result

some of the models available in models directory cannot be selected in run-me-nvidia.bat
thf-3 can be selected, but no model available.

Python ffprobe wrapper
Here

3.0.0.2a stores processed frames permanently. Therefore the processed part of a video can already be played while processing of the rest will continue. Also processing awake again with the PC coming out of energy savings mode. If that will be available in the final VEAI release that will be a great feature.

Also if there is a crash or if the user terminates the processing(pause) the video can still play. Later with the GUI app a resume/recover feature will be added to continue processing where it was left of.
Will add a do not sleep option to the filters so machine will not go to sleep.

5 Likes

This works well on my system, however ghq-6 results in significant artifacts in the form of halos around moving elements.

Indeed, same here (mac os)