> -----Original Message-----
> From: Raja Rathour via ffmpeg-devel <[email protected]>
> Sent: Thursday, December 25, 2025 6:33 PM
> To: [email protected]
> Cc: Raja Rathour <[email protected]>
> Subject: [FFmpeg-devel] [PATCH 1/3] avfilter/dnn_backend_torch: implement
> async execution
>
>
> Signed-off-by: Raja Rathour <[email protected]>
>
> This patch implements asynchronous model execution for the LibTorch
> backend in FFmpeg's DNN module.
>
> Key changes:
> - Integrated a worker thread and a pending queue to handle inference.
> - Prevents the main filter thread from blocking during model execution.
> - Aligns LibTorch backend behavior with the existing OpenVINO async
> implementation.
> - Improves overall throughput for deep learning filters using LibTorch.
>
> The implementation has been tested with various torch models to ensure
> stability and correct frame synchronization.
>
Thanks for the patch, is there PATCH 2/3 and 3/3?
_______________________________________________
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
