# Regarding the House Material Change
**URL:**
**Author:**  [@Jonas18369](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJonas18369)%5C
**Post date:** [June 12, 2025, 9:12am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fregarding-the-house-material-change%2F159027%2F1 "2025-06-12T09:12:35Z")
Hi, everyone.
Hope you are doing well.
Now I wanna make project that similar to [https://renovateai.app/](https://renovateai.app/).
If users upload house image and select material, change material according to the part.
But I’m not sure how should i do for that… which model should i use or what is best approach?
If anyone know about this, please let me know.
Thanks.
---
**Author:**  [@John6666](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJohn6666)%5C
**Post date:** [June 12, 2025, 9:26am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fregarding-the-house-material-change%2F159027%2F2 "2025-06-12T09:26:34Z")
There are two main approaches: using a standard diffusion model or using a diffusion model designed for replacement, such as the VTON series. (In the case below, the target is clothing…)
When using a standard diffusion model, techniques such as Inpainting, IP-Adapter, or ControlNet would likely be employed.
> **[yisol/IDM-VTON · Hugging Face](https://huggingface.co/yisol/IDM-VTON)**
>
> We’re on a journey to advance and democratize artificial intelligence through open source and open science.
> **[Inpainting](https://huggingface.co/docs/diffusers/using-diffusers/inpaint)**
>
> We’re on a journey to advance and democratize artificial intelligence through open source and open science.
> **[IP-Adapter](https://huggingface.co/docs/diffusers/main/en/using-diffusers/ip_adapter?usage=inpainting)**
>
> We’re on a journey to advance and democratize artificial intelligence through open source and open science.
> **[black-forest-labs/FLUX.1-Fill-dev · Hugging Face](https://huggingface.co/black-forest-labs/FLUX.1-Fill-dev)**
>
> We’re on a journey to advance and democratize artificial intelligence through open source and open science.
---
**Author:**  [@Jonas18369](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJonas18369)%5C
**Post date:** [June 12, 2025, 9:34am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fregarding-the-house-material-change%2F159027%2F3 "2025-06-12T09:34:14Z")
Thanks, @John6666
But could you explain in more detail?
---
**Author:**  [@John6666](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJohn6666)%5C
**Post date:** [June 12, 2025, 12:47pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fregarding-the-house-material-change%2F159027%2F4 "2025-06-12T12:47:45Z")
The overall process is the same for all cases. First, select the area of the object you want to replace. This can be done manually by the user. If you want to automate this step, use some kind of object detection model (there are many existing options such as YOLO or SAM).
Based on the selected area and the room image, create a mask image (a black-and-white image specifying the area to be redrawn) using image processing with Python’s OpenCV or similar tools.
When the room image and mask image are passed to the Diffusion model’s Inpaint pipeline, an image is output where only the masked area is redrawn as naturally as possible.
This is the basic flow.
This workflow assumes the use of a standard image generation model, so it is inefficient for the specific purpose, but it has the advantage of not requiring a dedicated model.
If a dedicated redrawing model for rooms exists, using that would be the simplest solution, but I am unsure if such a model exists.
> **[GitHub - yuqwu/Replace-Anything: A simple web application that lets you replace any...](https://github.com/yuqwu/Replace-Anything)**
>
> A simple web application that lets you replace any part of an image with an image generated based on your description.
> [@Object Removal in the the new Inpainting SDXL model?](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fobject-removal-in-the-the-new-inpainting-sdxl-model%2F53851)%3A
>
> Hi Everyone! I just started using it and the results are great! However, it has one big issue: It’s hard to remove a big object out of the input. Let’s say I want to empty a room, instead, it will keep throwing random asterisks into the input. Any ideas how to implement or setting the parameters to achieve this? EditPreview
> **[lrzjason/ObjectRemovalFluxFill · Hugging Face](https://huggingface.co/lrzjason/ObjectRemovalFluxFill)**
>
> We’re on a journey to advance and democratize artificial intelligence through open source and open science.
> **[Object To Object Replace - a Hugging Face Space by egg22314](https://huggingface.co/spaces/egg22314/object-to-object-replace)**
>
> Upload a main image and an object image to replace a specified object in the main image with the new object. Choose from a list of objects to replace and get the modified image as a result.
---
**Author:**  [@John6666](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJohn6666)%5C
**Post date:** [June 13, 2025, 5:33am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fregarding-the-house-material-change%2F159027%2F5 "2025-06-13T05:33:29Z")
I found an explanation by Diffusers’s developer.
> **[How to inpaint an existing object on top of another existing object? ·...](https://github.com/huggingface/diffusers/discussions/8374)**
>
> Hi there Above, I got this Nike shoe. My goal is simple. Take a photo of an actor/actress wearing a pair of shoes and in paint this Nike instead of what they are wearing. How can I do this in a way...