TTPlanet/HunyuanDiT_Controlnet

0
0
0
0
PhotographyRealisticGirlComposition Control
อัปเดตเมื่อ เผยแพร่ครั้งแรก:
Photography,Realistic,Girl,Controlnet,HunyuanDiT v1.2Image info
Photography,Realistic,Girl,Controlnet,HunyuanDiT v1.2Image info
Photography,Realistic,Girl,Controlnet,HunyuanDiT v1.2Image info
Photography,Realistic,Girl,Controlnet,HunyuanDiT v1.2Image info
Photography,Realistic,Girl,Controlnet,HunyuanDiT v1.2Image info
Photography,Realistic,Girl,Controlnet,HunyuanDiT v1.2Image info
Photography,Realistic,Girl,Controlnet,HunyuanDiT v1.2Image info
Photography,Realistic,Girl,Controlnet,HunyuanDiT v1.2Image info
Photography,Realistic,Girl,Controlnet,HunyuanDiT v1.2Image info
Photography,Realistic,Girl,Controlnet,HunyuanDiT v1.2Image info

Tencent HunyuanDiT Inpainting Controlnet

An amazing controlnet model which can provide you incredible consistency and creativity to the original image!

It is trained with various resolutions of images and masks to provide nice robustness to the base model. Now it can create a new image with the same style as the original one. Following are some examples to show the result! Comfyui inpainting preprocessor to combine the mask and image: https://github.com/TTPlanetPig/Comfyui_TTP_CN_Preprocessor

Let's have a game!

Which one is the original one? Select wisely.

The left one is No. 1, and the right one is No. 2.

This is an amazing controlnet model that can provide incredible consistency and creativity to the original image! The model is trained on various resolutions of images and masks to enhance the robustness of the base model. Now it can create new images with the same style as the original one. Below are some examples of the results! Comfyui repair preprocessor is used to combine masks and images: https://github.com/TTPlanetPig/Comfyui_TTP_CN_Preprocessor. Let's play a game! Which one is the original? Please choose wisely. The one on the left is No. 1, and the one on the right is No. 2.

imageNo1imageNo2

I have conducted my own test, and for most people, they will select No. 1 as the original image. However, No. 2 is the original one, and No. 1 is the image with inpainting control with the Hunyuan DiT model.

Other capabilities!

Recovery from 70% masked area but still maintaining consistency!

Example: Masked image:

The repaired image from the control model. Other features! Even with a masked area reaching 70%, consistency can still be maintained! Example: Masked image: a muscle man with blue eyes. Masked image: a man in a Batman suit with an eye cover.

Multiple-area inpainting with different prompts and understanding the depth of field correctly!

Masked 3 individual areas:

Multiple-area repair, with different prompts, and correct understanding of depth of field! Three independent areas are covered: An Asian man with a long sleeve T-shirt, in the background there is a yellow dog.

Multiple-area inpainting with high consistency!

imageNo1imageNo1imageNo2

Amazing outpainting!

imageNo1imageNo2

High-quality inpainting adapter to the original photo style or lighting!

Masked image:Description Text

imageNo1imageNo2

imageNo1imageNo2

In the end, enjoy the HunyuanDiT.

Join my QQ group if you appreciate my work: 294060503 or contact me on WeChat: tangtuanzhu

Finally, enjoy the HunyuanDiT experience. If you appreciate my work, please join my QQ group: 294060503, or contact me on WeChat: tangtuanzhu.

การอภิปราย

ที่นิยมมากที่สุด
|
ใหม่ที่สุด
ส่ง
เร็วๆ นี้
ดาวน์โหลด
(0.00KB)
รายละเอียด
ประเภท
จำนวนการสร้างรูปภาพออนไลน์
0
ดาวน์โหลด
0

แกลเลอรี

ที่นิยมมากที่สุด
|
ใหม่ที่สุด