Faceswap Forum

I figured out to combine extracted frames from different video sources + individual images into single folder for extraction/masking/alignments from the Extraction FAQ, I think it was. Like you, I saw it would be added cycles to perform the same process across each source, when in the end the applic...

I tried [mention]MaxHunter[/mention] 's idea and split up the source video, but it turns out that the Convert step for each section also hit unresolvable pauses in every try. So, Convert may have more issues in dealing with certain videos compared to other faceswap tools which showed no such limitat...

There's definitely something wrong in Convert's perceptions of how long the video happens to be. I copied the prior alignments file and extracted alignment .PNGs to another folder, then deleted the initial 25000 PNGs and removed those frames from the alignments file. It was a guess that maybe fewer ...

That's interesting - in this case, I am actually using a 26 minute portion of a longer video and thought that maybe there was some sort of "limit" being reached, but further analysis showed that the Convert process started to encounter errors and finally stopped without explanation: ------...

If it's hit 100% and all the frames are in the output folder, it's finished. I have never had a problem with convert getting stuck though. So, I had a closer look and it seems that Convert is saying 100% but has not actually processed the entire video. That's why the Stop button remains on, apparen...

Thanks! I clicked the button and the output files were usable for creating a final, converted movie file.

I'll see if I can recreate with a smaller project and send across, if so.

Hello, This is my first Convert attempt and my question is if Convert's "Stop" button (shown below) implies that it is still processing the outputs, because it's been four hours since it generated converted frames in the output folder and I've seen no obvious progress since that time. In T...

Allow growth is, sadly, also going to give you a bit of a performance hit. Sadly it seems unavoidable in your case, but it basically means your VRAM will be more fragmented, so not as fast. Understood, though from what I am reading here: https://www.tensorflow.org/guide/gpu The first option is to t...

Hoping this is helpful to share. On my Windows 10 Pro system with GPU processing enabled in Faceswap, I happened to notice that for Dlight and Villain models (will be testing others), when Hardware-accelerated GPU scheduling was enabled under Windows, my NVidia 30x0 cards performed 16% EGs/sec faste...

Short update, still testing various options: Because the reported issue seemed to be related to a problem allocating VRAM when both GPU cards were enabled, I tried checking the Train -> Settings -> Configure Settings -> Train -> Network -> Allow Growth option, because the Help says it prevents Tenso...

Yeah, it's weird, but thanks for looking. Doing a diff of the two Trace-level logs (between 1xGPU and 2xGPU) the major difference is that when the 1xGPU case successfully loads the model: MainProcess _training_0 _base _load INFO Loaded model from disk: 'D:\faceswap\sourceA\modelAB\dlight.h5' the 2xG...

Thanks for the idea, unfortunately everything fails with what appears to be a memory allocation error when two GPUs are enabled for training. Tried deleting the existing model folder and letting the Training module recreate a new model for every test run Tried deselecting both Initializers before a ...

Thanks for offering to look at what's happening! I should probably confirm that this system is regularly used to render using both GPUs, so their working together with other software packages is at least something I know works today. I've tried to anonymize the data and sent you a PM with the downlo...

Hello, First off, thanks for this really interesting toolset - this is the first time I've begun trying it out. System basics : OS: Windows 10 Pro motherboard: ASUS Pro WS X570-ACE RAM: 64GG (DDR4) GPUs: 2 x ASUS ROG Strix RTX3090-024G FaceSwap version: updated today via the FS menu option (06/07/20...