Kadir Nar
Kadir Nar
> 6gb vram ile 1660 Ti'm var, points_per_batch=16 parametresini 2 ile bile denedim ve hala aynı miktarda bellek kullanıyorum. > > Ayrıca görüntü çözünürlüğünü orijinal görüntünün dörtte birine kadar düşürmeyi...
I added pip and video support. You were to try: https://github.com/kadirnar/segment-anything-video
> Because the diff doesn't show inline, here's the new function for anyone who is interested: > > ``` > def show_anns(anns): > if len(anns) == 0: > return >...
Hi, you need to reduce the parameter values. Normally you need 6-12gb vram. I ran gtx 1050ti 2gb vram as well(for image) https://github.com/kadirnar/segment-anything-video
> Hello, that's really awesome results. Did you run it with cuda for all the frames ? If so, is it on your own environment ? I'm also trying to...
Can you share the image file?
> Similar to high res inpainting in SD webui, would this be possible? I want to add this feature to my demo application. But I couldn't find the codes. Can...
Can you do it? I don't want to change the folders structure.
> Hi @csndylim - Not currently. But I am working on adding support for running via `intel-transformers` or default `ONNX-runtime` this might take a bit tho. https://github.com/MiscellaneousStuff/openai-whisper-cpu ?
@CapsAdmin Would you look at this? I get bad output when inpainting.