Stable Diffusion ComfyUI With An Awesome AI Background Removal - InSPyReNet
HTML-код
- Опубликовано: 5 окт 2024
- We'll be exploring the InSPyReNet background removal method and comparing it to other popular methods like SAM and U2Net. We'll walk through the installation process and demonstrate how it performs in removing backgrounds from images.
ComfyUI With Florence 2 Vision LLM
• ComfyUI With Florence ...
Segment Anything And Mask With Image Or Video
• Stable Diffusion Use S...
Freebie Workflow Here : www.patreon.co...
If you're tired of struggling with background removal and are looking for the best method out there, you've come to the right place. We'll show you step-by-step how to install and use the InSPyReNet custom node in ComfyUI. You'll be amazed at how smoothly it removes backgrounds while retaining important details like hair and clothing.
In this video, we'll also do a side-by-side comparison of InSPyReNet with other background removal methods, highlighting its advantages and showcasing its superior results. You'll see how InSPyReNet outperforms other methods when it comes to preserving intricate features and maintaining the integrity of the subject.
Throughout the video, we'll provide valuable insights and tips on optimizing your background removal workflow. Whether you're a content creator, filmmaker, or animator, having a reliable and efficient background removal tool is essential. InSPyReNet might just be the game-changer you've been looking for.
So, if you're ready to take your background removal skills to the next level, join us in this exciting video and discover the power of InSPyReNet. Don't forget to like, comment, and subscribe to our channel for more awesome content like this. Let's dive in!
If You Like tutorial like this, You Can Support Our Work In Patreon:
/ aifuturetech
Discord : / discord
Hi
I'm the author of the node
I liked your video, and I'm glad that the node was helpful
Glad to hear that!
Nice work on the node, thanks very much.
dude thx so much for sharing, ya all da best man
This bg removal model do awesome work！
always looking for better ways to remove backgrounds
Well, I try it. And it removed background well clean.
Its smart enough to know an object that is holding by a character. The object on the hand have not removed, which is what I want.
Looks like you add the green backdrops at the end of this video, the purpose of this, use for an object stitch into an animation, right?
You got it😎👍 And you know why I used the green bg. Well it can be image bg if use it for image only.
thanks for the video, looks like a better bgrem than the older ones. but, why would you want to add a green bg after you removed the bg? just save a png sequence with alpha, the point of green screen in real video footage is to easily remove the bg. if you have transparent bg already you don’t need to add green bg so you would need to remove it AGAIN
Such a great tutotrial!
Can you please make a tutorial how to do use it with VIDEO ?
Thanks ahead💙
About "FOND VERT". (green background)
First of all, thank you so much for everything you're teaching me
with your super clear and detailed videos. It's great to be able to ,
thanks to ComfyUI, build a change of background in 2 steps (step 1:
green background, step 2: inlay in a new background).
it's only possible with photos, isn't it? The next big thing will be when
change the background of a video! whouahhoo , with a minimum of conditions of course
conditions of course, a clearly identifiable central character (e.g. a man or woman
man or woman walking or running in the street).
To your knowledge, does this already exist? Or have you
have you heard of any first steps in this development of a background change for a video?
a video? Very interested in the subject.
Thanks, does it work with video too? if yes what node should i use for input and output.?
Love ya bro! thanks!
You're welcome
Does it also remove glass objects with transparent BG?
Its the best way has an option on generator is "no background" directly. And supported PNG.
Great work! I tried InSPyReNet in comfyui. It can show the mask generated but the background of the image is not removed. Any ideas why?
This is really good
How about : taking photoshop bash comps and making variations that are professional.
So much of the output is low quality?
Why not 16 bit exr finals ?
seems to work great for images but I don't get it to work with videos. It gives me the same output as the input and the background was not removed. Any ideas?
I did use Video Combine node and run with this bg removal. It works on my system. Maybe before input add a nide call ImageList To Image Batch , and do a reverse in output connection. Try it.
See my recent Mimic motion video , i have use this as well.
You should search segment anything version 2 nodes for comfyui. Segment anything 2 is a Facebook model that is specifically designed to mask and track objects on videos.
Something like this can remove a single character from an image featuring a group of characters?
Yeah. Keyword on “something like this”. It can do it but will require quite a bit of work to modify
SEGs can pick based on people it finds.
@@alecubudulecu nope, this one is just bg removeal. Selective object can be use Segment Anything, doesn't need lots of work or mask.
top
How can i add this 2:27 thing? #36 essentialblabla
install custom node "ComfyUI Essentials"